Tag Archives: serverless

Introducing the AWS Integrated Application Test Kit (IATK)

2023-11-16 James Beswick

Post Syndicated from James Beswick original https://aws.amazon.com/blogs/compute/aws-integrated-application-test-kit/

This post is written by Dan Fox, Principal Specialist Solutions Architect, and Brian Krygsman, Senior Solutions Architect.

Today, AWS announced the public preview launch of the AWS Integrated Application Test Kit (IATK). AWS IATK is a software library that helps you write automated tests for cloud-based applications. This blog post presents several initial features of AWS IATK, and then shows working examples using an example video processing application. If you are getting started with serverless testing, learn more at serverlessland.com/testing.

Overview

When you create applications composed of serverless services like AWS Lambda, Amazon EventBridge, or AWS Step Functions, many of your architecture components cannot be deployed to your desktop, but instead only exist in the AWS Cloud. In contrast to working with applications deployed locally, these types of applications benefit from cloud-based strategies for performing automated tests. For its public preview launch, AWS IATK helps you implement some of these strategies for Python applications. AWS IATK will support other languages in future launches.

Locating resources for tests

When you write automated tests for cloud resources, you need the physical IDs of your resources. The physical ID is the name AWS assigns to a resource after creation. For example, to send requests to Amazon API Gateway you need the physical ID, which forms the API endpoint.

If you deploy cloud resources in separate infrastructure as code stacks, you might have difficulty locating physical IDs. In CloudFormation, you create the logical IDs of the resources in your template, as well as the stack name. With IATK, you can get the physical ID of a resource if you provide the logical ID and stack name. You can also get stack outputs by providing the stack name. These convenient methods simplify locating resources for the tests that you write.

Creating test harnesses for event driven architectures

To write integration tests for event driven architectures, establish logical boundaries by breaking your application into subsystems. Your subsystems should be simple enough to reason about, and contain understandable inputs and outputs. One useful technique for testing subsystems is to create test harnesses. Test harnesses are resources that you create specifically for testing subsystems.

For example, an integration test can begin a subsystem process by passing an input test event to it. IATK can create a test harness for you that listens to Amazon EventBridge for output events. (Under the hood, the harness is composed of an EventBridge Rule that forwards the output event to Amazon Simple Queue Service.) Your integration test then queries the test harness to examine the output and determine if the test passes or fails. These harnesses help you create integration tests in the cloud for event driven architectures.

Establishing service level agreements to test asynchronous features

If you write a synchronous service, your automated tests make requests and expect immediate responses. When your architecture is asynchronous, your service accepts a request and then performs a set of actions at a later time. How can you test for the success of an activity if it does not have a specified duration?

Consider creating reasonable timeouts for your asynchronous systems. Document timeouts as service level agreements (SLAs). You may decide to publish your SLAs externally or to document them as internal standards. IATK contains a polling feature that allows you to establish timeouts. This feature helps you to test that your asynchronous systems complete tasks in a timely manner.

Using AWS X-Ray for detailed testing

If you want to gain more visibility into the interior details of your application, instrument with AWS X-Ray. With AWS X-Ray, you trace the path of an event through multiple services. IATK provides conveniences that help you set the AWS X-Ray sampling rate, get trace trees, and assert for trace durations. These features help you observe and test your distributed systems in greater detail.

Learn more about testing asynchronous architectures at aws-samples/serverless-test-samples.

Overview of the example application

To demonstrate the features of IATK, this post uses a portion of a serverless video application designed with a plugin architecture. A core development team creates the primary application. Distributed development teams throughout the organization create the plugins. One AWS CloudFormation stack deploys the primary application. Separate stacks deploy each plugin.

Communications between the primary application and the plugins are managed by an EventBridge bus. Plugins pull application lifecycle events off the bus and must put completion notification events back on the bus within 20 seconds. For testing, the core team has created an AWS Step Functions workflow that mimics the production process by emitting properly formatted example lifecycle events. Developers run this test workflow in development and test environments to verify that their plugins are communicating properly with the event bus.

The following demonstration shows an integration test for the example application that validates plugin behavior. In the integration test, IATK locates the Step Functions workflow. It creates a test harness to listen for the event completion notification to be sent by the plugin. The test then runs the workflow to begin the lifecycle process and start plugin actions. Then IATK uses a polling mechanism with a timeout to verify that the plugin complies with the 20 second service level agreement. This is the sequence of processing:

The integration test starts an execution of the test workflow.
The workflow puts a lifecycle event onto the bus.
The plugin pulls the lifecycle event from the bus.
When the plugin is complete, it puts a completion event onto the bus.
The integration test polls for the completion event to determine if the test passes within the SLA.

Deploying and testing the example application

Follow these steps to review this application, build it locally, deploy it in your AWS account, and test it.

Downloading the example application

Open your terminal and clone the example application from GitHub with the following command or download the code. This repository also includes other example patterns for testing serverless applications.
```
git clone https://github.com/aws-samples/serverless-test-samples
```
The root of the IATK example application is in python-test-samples/integrated-application-test-kit. Change to this directory:
```
cd serverless-test-samples/python-test-samples/integrated-application-test-kit
```

Reviewing the integration test

Before deploying the application, review how the integration test uses the IATK by opening plugins/2-postvalidate-plugins/python-minimal-plugin/tests/integration/test_by_polling.py in your text editor. The test class instantiates the IATK at the top of the file.

iatk_client = aws_iatk.AwsIatk(region=aws_region)

In the setUp() method, the test class uses IATK to fetch CloudFormation stack outputs. These outputs are references to deployed cloud components like the plugin tester AWS Step Functions workflow:

stack_outputs = self.iatk_client.get_stack_outputs(
    stack_name=self.plugin_tester_stack_name,
    output_names=[
        "PluginLifecycleWorkflow",
        "PluginSuccessEventRuleName"
    ],
)

The test class attaches a listener to the default event bus using an Event Rule provided in the stack outputs. The test uses this listener later to poll for events.

add_listener_output = self.iatk_client.add_listener(
    event_bus_name="default",
    rule_name=self.existing_rule_name
)

The test class cleans up the listener in the tearDown() method.

self.iatk_client.remove_listeners(
    ids=[self.listener_id]
)

Once the configurations are complete, the method test_minimal_plugin_event_published_polling() implements the actual test.

The test first initializes the trigger event.

trigger_event = {
    "eventHook": "postValidate",
    "pluginTitle": "PythonMinimalPlugin"
}

Next, the test starts an execution of the plugin tester Step Functions workflow. It uses the plugin_tester_arn that was fetched during setUp.

self.step_functions_client.start_execution(
    stateMachineArn=self.plugin_tester_arn,
    input=json.dumps(trigger_event)
)

The test polls the listener, waiting for the plugin to emit events. It stops polling once it hits the SLA timeout or receives the maximum number of messages.

poll_output = self.iatk_client.poll_events(
    listener_id=self.listener_id,
    wait_time_seconds=self.SLA_TIMEOUT_SECONDS,
    max_number_of_messages=1,
)

Finally, the test asserts that it receives the right number of events, and that they are well-formed.

self.assertEqual(len(poll_output.events), 1)
self.assertEqual(received_event["source"], "video.plugin.PythonMinimalPlugin")
self.assertEqual(received_event["detail-type"], "plugin-complete")

Installing prerequisites

You need the following prerequisites to build this example:

An AWS account
The AWS Serverless Application Model (AWS SAM) CLI with credentials that can manage AWS resources
Python 3.11
Node.js 18.x
Docker (optional, but recommended for building Python applications with AWS SAM)

Build and deploy the example application components

Use AWS SAM to build and deploy the plugin tester to your AWS account. The plugin tester is the Step Functions workflow shown in the preceding diagram. During the build process, you can add the --use-container flag to the build command to instruct AWS SAM to create the application in a provided container. You can accept or override the default values during the deploy process. You will use “Stack Name” and “AWS Region” later to run the integration test.
```
cd plugins/plugin_tester # Move to the plugin tester directory

sam build --use-container # Build the plugin tester
```

Deploy the tester:

sam deploy --guided # Deploy the plugin tester

Once the plugin tester is deployed, use AWS SAM to deploy the plugin.

cd ../2-postvalidate-plugins/python-minimal-plugin # Move to the plugin directory

sam build --use-container # Build the plugin

Deploy the plugin:
```
sam deploy --guided # Deploy the plugin
```

Running the test

You can run tests written with IATK using standard Python test runners like unittest and pytest. The example application test uses unittest.

1. Use a virtual environment to organize your dependencies. From the root of the example application, run:
```
python3 -m venv .venv # Create the virtual environment
source .venv/bin/activate # Activate the virtual environment
```
2. Install the dependencies, including the IATK:
```
cd tests 
pip3 install -r requirements.txt
```
3. Run the test, providing the required environment variables from the earlier deployments. You can find correct values in the samconfig.toml file of the plugin_tester directory.
```
cd integration

PLUGIN_TESTER_STACK_NAME=video-plugin-tester \
AWS_REGION=us-west-2 \
python3 -m unittest ./test_by_polling.py
```

You should see output as unittest runs the test.

Open the Step Functions console in your AWS account, then choose the PluginLifecycleWorkflow-<random value> workflow to validate that the plugin tester successfully ran. A recent execution shows a Succeeded status:

Review other IATK features

The example application includes examples of other IATK features like generating mock events and retrieving AWS X-Ray traces.

Cleaning up

Use AWS SAM to clean up both the plugin and the plugin tester resources from your AWS account.

Delete the plugin resources:

cd ../.. # Move to the plugin directory
sam delete # Delete the plugin

Delete the plugin tester resources:

cd ../../plugin_tester # Move to the plugin tester directory
sam delete # Delete the plugin tester

The temporary test harness resources that IATK created during the test are cleaned up when the tearDown method runs. If there are problems during teardown, some resources may not be deleted. IATK adds tags to all resources that it creates. You can use these tags to locate the resources then manually remove them. You can also add your own tags.

Conclusion

The AWS Integrated Application Test Kit is a software library that provides conveniences to help you write automated tests for your cloud applications. This blog post shows some of the features of the initial Python version of the IATK.

To learn more about automated testing for serverless applications, visit serverlessland.com/testing. You can also view code examples at serverlessland.com/testing/patterns or at the AWS serverless-test-samples repository on GitHub.

For more serverless learning resources, visit Serverless Land.

Triggering AWS Lambda function from a cross-account Amazon Managed Streaming for Apache Kafka

2023-11-16 Marcia Villalba

Post Syndicated from Marcia Villalba original https://aws.amazon.com/blogs/compute/triggering-aws-lambda-function-from-a-cross-account-amazon-managed-streaming-for-apache-kafka/

This post is written by Subham Rakshit, Senior Specialist Solutions Architect, and Ismail Makhlouf, Senior Specialist Solutions Architect.

Many organizations use a multi-account strategy for stream processing applications. This involves decomposing the overall architecture into a single producer account and many consumer accounts. Within AWS, in the producer account, you can use Amazon Managed Streaming for Apache Kafka (Amazon MSK), and in their consumer accounts have AWS Lambda functions for event consumption. This blog post explains how you can trigger Lambda functions from a cross-account Amazon MSK cluster.

The Lambda event sourcing mapping (ESM) for Amazon MSK continuously polls for new events from the Amazon MSK cluster, aggregates them into batches, and then triggers the target Lambda function. The ESM for Amazon MSK functions as a serverless set of Kafka consumers that ensures that each event is processed at least once. Additionally, events are processed in the same order they are received within each Kafka partition. In addition, the ESM batches the stream of data and filters the events based on configured logic.

Overview

Amazon MSK supports two different deployment types: provisioned and serverless. Triggering a Lambda function from a cross-account Amazon MSK cluster is only supported with a provisioned cluster deployed within the same Region. To facilitate this functionality, Amazon MSK uses multi-VPC private connectivity, powered by AWS PrivateLink, which simplifies connecting Kafka consumers hosted in different AWS accounts to an Amazon MSK cluster.

The following diagram illustrates the architecture of this example:

The architecture is divided in two parts: the producer and the consumer.

In the producer account, you have the Amazon MSK cluster with multi-VPC connectivity enabled. Multi-VPC connectivity is only available for authenticated Amazon MSK clusters. Cluster policies are required to grant permissions to other AWS accounts, allowing them to establish private connectivity to the Amazon MSK cluster. You can delegate permissions to relevant roles or users. When combined with AWS Identity and Access Management (IAM) client authentication, cluster policies offer fine-grained control over Kafka data plane permissions for connecting applications.

In the consumer account, you have the Lambda ESM for Amazon MSK and the managed VPC connection deployed within the same VPC. The managed VPC connection allows private connectivity from the consumer application VPC to the Amazon MSK cluster. The Lambda ESM for Amazon MSK connects to the cross-account Amazon MSK cluster via IAM authentication. It also supports SASL/SCRAM, and mutual TLS (mTLS) authenticated clusters. The ESM receives the event from the Kafka topic and invokes the Lambda function to process it.

Deploying the example application

To set up the Lambda function trigger from a cross-account Amazon MSK cluster as the event source, follow these steps. The AWS CloudFormation templates for deploying the example are accessible in the GitHub repository.

As a part of this example, some sample data is published using the Kafka console producer and Lambda processes these events and writes to Amazon S3.

Pre-requisites

For this example, you need two AWS accounts. This post uses the following naming conventions:

Producer (for example, account No: 1111 1111 1111): Account that hosts the Amazon MSK cluster and Kafka client instance.
Consumer (for example, account No: 2222 2222 2222): Account that hosts the Lambda function and consumes events from Amazon MSK.

To get started:

Clone the repository locally:
git clone https://github.com/aws-samples/lambda-cross-account-msk.git
Set up the producer account: you must configure the VPC networking, deploy the Amazon MSK cluster, and a Kafka client instance to publish data. To do this, deploy the CloudFormation template producer-account.yaml from the AWS console and take note of the MSKClusterARN from the CloudFormation outputs tab.
Set up the consumer account: To set up the consumer account, you need the Lambda function, IAM role used by the Lambda function, and S3 bucket receiving the data. For this, deploy the CloudFormation template consumer-account.yaml from the AWS console with the input parameter MSKAccountId, that is the producer AWS account ID (for example, account Id: 1111 1111 1111). Note the LambdaRoleArn from the CloudFormation outputs tab.

Setting up multi-VPC connectivity in the Amazon MSK cluster

Once the accounts are created, you must enable connectivity between them. By enabling multi-VPC private connectivity in the Amazon MSK cluster, you set up the network connection to allow the cross-account consumers to connect to the cluster.

In the producer account, navigate to the Amazon MSK console.
Choose producer-cluster, and go to the Properties tab.
Scroll to Networking settings, choose Edit, and select Turn on multi-VPC connectivity. This takes some time, then appears as follows.
Add the necessary cluster policy to allow cross-account consumers to connect to Amazon MSK. In the producer account, deploy the CloudFormation template producer-msk-cluster-policy.yaml from the AWS console with the following input parameters:

- MSKClusterArn – Amazon Resource Name (ARN) of the Amazon MSK cluster in producer account. Find this information in the CloudFormation output of producer-account.yaml.
- LambdaRoleArn – ARN of the IAM role attached to the Lambda function in the consumer account. Find this information in the CloudFormation output of consumer-account.yaml.
- LambdaAccountId – Consumer AWS account ID (for example, account Id: 2222 2222 2222).

Creating a Kafka topic in Amazon MSK and publishing events

In the producer account, navigate to the Amazon MSK console. Choose the Amazon MSK cluster named producer-cluster. Choose View client information to show the bootstrap server.

The CloudFormation template also deploys a Kafka client instance to create topics and publish events.

To access the client, go to the Amazon EC2 console and choose the instance producer-KafkaClientInstance1. Connect to EC2 instance with Session Manager:

sudo su - ec2-user
#Set MSK Broker IAM endpoint
export BS=<<Provide IAM bootstrap address here>>

You must use the single-VPC Private endpoint for the Amazon MSK cluster and not the multi-VPC private endpoint, as you are going to publish events from a Kafka console producer from the producer account.

Run these scripts to create the customer topic and publish sample events in the topic:

./kafka_create_topic.sh
./kafka_produce_events.sh

Creating a managed VPC connection in the consumer account

To establish a connection to the Amazon MSK cluster in the producer account, you must create a managed VPC connection in the consumer account. Lambda communicates with cross-account Amazon MSK through this managed VPC connection.

For detailed setup steps, read the Amazon MSK managed VPC connection documentation.

Configuring the Lambda ESM for Amazon MSK

The final step is to set up the Lambda ESM for Amazon MSK. Setting up the ESM enables you to connect to the Amazon MSK cluster in the producer account via the managed VPC endpoint. This allows you to trigger the Lambda function to process the data produced from the Kafka topic:

In the consumer account, go to the Lambda console.
Open the Lambda function msk-lambda-cross-account-iam.
Go to the Configuration tab, select Triggers, and choose Add Trigger.
For Trigger configuration, select Amazon MSK.

To configure this trigger:

Select the shared Amazon MSK cluster. This automatically defaults to the IAM authentication that is used to connect to the cluster.
By default, the Active trigger check box is enabled. This ensures that the trigger is in the active state after creation. For the other values:
1. Keep the Batch size default to 100.
2. Change the Starting Position to Trim horizon.
3. Set the Topic name as customer.
4. Set the Consumer Group ID as msk-lambda-iam.

Scroll to the bottom and choose Add. This starts creating the Amazon MSK trigger, which takes several minutes. After creation, the state of the trigger shows as Enabled.

Verifying the output on the consumer side

The Lambda function receives the events and writes them in an S3 bucket.

To validate that the function is working, go to the consumer account and navigate to the S3 console. Search for the cross-account-lambda-consumer-data-<<REGION>>-<<AWS Account Id>> bucket. In the bucket, you see the customer-data-<<datetime>>.csv files.

Cleaning up

You must empty and delete the S3 bucket, managed VPC connection, and the Lambda ESM for Amazon MSK manually from the consumer account. Next, delete the CloudFormation stacks from the AWS console from both the producer and consumer accounts to remove all other resources created as a part of the example.

Conclusion

With Lambda and Amazon MSK, you can now build a decentralized application distributed across multiple AWS accounts. This post shows how you can set up Amazon MSK as an event source for cross-account Lambda functions and also walks you through the configuration required in both producer and consumer accounts.

For further reading on AWS Lambda with Amazon MSK as an event source, visit the documentation.

For more serverless learning resources, visit Serverless Land.

Introducing AWS Step Functions redrive to recover from failures more easily

2023-11-15 Benjamin Smith

Post Syndicated from Benjamin Smith original https://aws.amazon.com/blogs/compute/introducing-aws-step-functions-redrive-a-new-way-to-restart-workflows/

Developers use AWS Step Functions, a visual workflow service to build distributed applications, automate IT and business processes, and orchestrate AWS services with minimal code.

Step Functions redrive for Standard Workflows allows you to redrive a failed workflow execution from its point of failure, rather than having to restart the entire workflow. This blog post explains how to use the new redrive feature to skip unnecessary workflow steps and reduce the cost of redriving failed workflows.

Handling workflow errors

Any workflow state can encounter runtime errors. Errors happen for various reasons, including state machine definition issues, task failures, incorrect permissions, and exceptions from downstream services. By default, when a state reports an error, Step Functions causes the workflow execution to fail. Step Functions allows you to handle errors by retrying, catching, and falling back to a defined state.

Now, you can also redrive the workflow from the failed state, skipping the successful prior workflow steps. This results in faster workflow completion and lower costs. You can only redrive a failed workflow execution from the step where it failed using the same input as the last non-successful state. You cannot redrive a failed workflow execution using a state machine definition that is different from the initial workflow execution.

Choosing between retry and redrive

Use the retry mechanism for transient issues such as network connectivity problems or momentary service unavailability You can configure the number of retries, along with intervals and back-off rates, providing the workflow with multiple attempts to complete a task successfully.

In scenarios where the underlying cause of an error requires longer investigation or resolution time, redrive becomes a valuable tool. Consider a situation where a downstream service experiences extended downtime or manual intervention is needed, such as updating a database or making code changes to a Lambda function. In these cases, being able to redrive the workflow can give you time to address the root cause before resuming the workflow execution.

Combining retry and redrive

Adopt a hybrid strategy that combines retry and redrive mechanisms:

Retry mechanism: Configure an initial set of retries for automatically resolvable errors. This ensures that transient issues are promptly addressed, and the workflow proceeds without unnecessary delays.
Error catching and redrive: If the retry mechanism exhausts without success, allow the state to fail and use the redrive feature to restart the workflow from the last non-successful state. This approach allows for intervention where errors persist or require external actions.

Reducing costs

AWS charges for Standard Workflows based on the number of state transitions required to run a workload. Step Functions counts a state transition each time a step of your workflow runs. Step Functions charges for the total number of state transitions across state machines, including retries. The cost is $0.025 per 1,000 state transitions. This means that reducing the number of state transitions reduces the cost of running your Standard Workflows.

If a workflow has many steps, includes parallel or map states, or is prone to errors that require frequent re-runs, this new feature reduces the costs incurred. You pay only for each state transition after the failed state and those costs for every downstream service invoked as part of the re-run.

The following example explains the cost implications of retrying a workflow that has failed, with and without redrive. In this example, a Step Functions workflow orchestrates Amazon Transcribe to generate a text transcription from an .mp4 file.

Since the failed state occurs towards the end of this workflow, the redrive execution does not run the successful states, reducing the overall successful completion time. If this workflow were to fail regularly, the reduction in transitions and execution duration becomes increasingly valuable.

The first time this workflow runs, the final state, which uses an AWS Lambda function to make an HTTP request fails with an IAM error. This is because the workflow does not have the required permissions to invoke the Lambda function. After granting the required permissions to the workflow’s execution role, redrive to continue the workflow from the failed state.

After the redrive, Step Functions workflow reports a different failure. This time it is related to the configuration of the Lambda function. This is an example of a downstream failure that does not require an update to my workflow definition.

After resolving the Lambda configuration issue and redriving the workflow, the execution completes successfully. The following image shows the execution details, including the number of redrives, the total state transitions, and the last redrive time:

Getting started with redrive

Redrive works for Standard Workflows only. You can redrive a workflow from its failed step programmatically, via the AWS CLI or AWS SDK, or using the Step Functions console, which provides a visual operator experience:

From the Step Functions console, select the failed workflow you want to redrive, and choose Redrive.
A modal appears with the execution details. Choose Redrive execution.

The state to redrive from, the workflow definition, and the previous input are immutable.

To redrive a workflow execution programmatically from its point of failure, call the new Redrive Execution API action. The same workflow execution starts from the last non-successful state and uses the same input as the last non-successful state from the initial failed workflow execution.

Programmatically catching failed workflow executions to redrive

Step Functions can process workloads autonomously, without the need for human interaction, or can include intervention from a user by implementing the .waitForTastToken pattern.

Redrive is for unhandled and unexpected errors only. Handling errors within a workflow using the built-in mechanisms for catch, retry, and routing to a Fail state, does not permit the workflow to redrive. However, it is possible to detect in near real-time when a workflow has failed, and programmatically redrive. When a workflow fails, it emits an event onto the Amazon EventBridge default event bus. The event looks like the following JSON object:

There are four new key/values pairs in this event:

"redriveCount": 0, 
"redriveDate": null, 
"redriveStatus": "REDRIVABLE", 
"redriveStatusReason": null,

The redrive count shows how many times the workflow has previously been redriven. The redrive status shows if the failed workflow is eligible for redrive execution.

To programmatically redrive the workflow from the failed state. Create a rule that pattern matches this event, and route the event onto a target service to handle the error. The target service uses the new States.RedriveExecution API to redrive the workflow.

Download and deploy the previous pattern from this example on serverlessland.com.

In the following example, the first state sends a post request to an API endpoint. If the request fails due to network connectivity or latency issues, the state retries. If the retry fails, then Step Functions emits a ` Step Functions Execution Status Change event onto the EventBridge default event bus. An EventBridge rule routes this event to a service where you can rectify this error and then redrive the task using the Step Functions API.

The new redrive feature also supports the distributed map state.

Redrive for express child workflow executions

For failed child workflow executions that are Express Workflows within a Distributed Map, the redrive capability ensures a seamless restart from the beginning of the child workflow. This allows you to resolve issues that are specific to individual iterations without restarting the entire map.

Redrive for standard child workflow executions

For failed child workflow executions within a Distributed Map that are Standard Workflows, the redrive feature functions in the same way in standalone Standard Workflows. You can restart the failed iteration from its point of failure, skipping unnecessary steps that have already successfully executed.

Conclusion

Step Functions redrive for Standard Workflows allows you to redrive a failed workflow execution from its point of failure rather than having to restart the entire workflow. This results in faster workflow completion and lower costs for processing failed executions. This is because it minimizes the number of state transitions and downstream service invocations.

Visit the Serverless Workflows Collection to browse the many deployable workflows to help build your serverless applications.

Clean up your Excel and CSV files without writing code using AWS Glue DataBrew

2023-11-15 Ismail Makhlouf

Post Syndicated from Ismail Makhlouf original https://aws.amazon.com/blogs/big-data/clean-up-your-excel-and-csv-files-without-writing-code-using-aws-glue-databrew/

Managing data within an organization is complex. Handling data from outside the organization adds even more complexity. As the organization receives data from multiple external vendors, it often arrives in different formats, typically Excel or CSV files, with each vendor using their own unique data layout and structure. In this blog post, we’ll explore a solution that streamlines this process by leveraging the capabilities of AWS Glue DataBrew.

DataBrew is an excellent tool for data quality and preprocessing. You can use its built-in transformations, recipes, as well as integrations with the AWS Glue Data Catalog and Amazon Simple Storage Service (Amazon S3) to preprocess the data in your landing zone, clean it up, and send it downstream for analytical processing.

In this post, we demonstrate the following:

Extracting non-transactional metadata from the top rows of a file and merging it with transactional data
Combining multi-line rows into single-line rows
Extracting unique identifiers from within strings or text

Solution overview

For this use case, imagine you’re a data analyst working at your organization. The sales leadership have requested a consolidated view of the net sales they are making from each of the organization’s suppliers. Unfortunately, this information is not available in a database. The sales data comes from each supplier in layouts like the following example.

However, with hundreds of resellers, manually extracting the information at the top is not feasible. Your goal is to clean up and flatten the data into the following output layout.

To achieve this, you can use pre-built transformations in DataBrew to quickly get the data in the layout you want.

Prerequisites

For this walkthrough, you should have the following prerequisites:

An AWS account.
An AWS Identity and Access Management (IAM) role with permissions for Amazon S3 and DataBrew. For more information, refer to Setting up AWS Identity and Access Management (IAM) permissions.

Connect to the dataset

The first thing we need to do is upload the input dataset to Amazon S3. Create an S3 bucket for the project and create a folder to upload the raw input data. The output data will be stored in another folder in a later step.

Next, we need to connect DataBrew to our CSV file. We create what we call a dataset, which
is an artifact that points to whatever data source we will be using. Navigate to “Datasets” on
the left hand menu.

Ensure the Column header values field is set to Add default header. The input CSV has an irregular format, so the first row will not have the needed column values.

Create a project

To create a new project, complete the following steps:

On the DataBrew console, choose Projects in the navigation pane.
Choose Create project.
For Project name, enter FoodMartSales-AllUpProject.
For Attached recipe, choose Create new recipe.
For Recipe name, enter FoodMartSales-AllUpProject-recipe.
For Select a dataset, select My datasets.
Select the FoodMartSales-AllUp dataset.
Under Permissions, for Role name, choose the IAM role you created as a prerequisite or create a new role.
Choose Create project.

After the project is opened, an interactive session is created where you can author transformations on a sample of the data.

Extract non-transactional metadata from within the contents of the file and merge it with transactional data

In this section, we consider data that has metadata on the first few rows of the file, followed by transactional data. We walk through how to extract data relevant to the whole file from the top of the document and combine it with the transactional data into one flat table.

Extract metadata from the header and remove invalid rows

Complete the following steps to extract metadata from the header:

Choose Conditions and then choose IF.
For Matching conditions, choose Match all conditions.
For Source, choose Value of and Column_1.
For Logical condition, choose Is exactly.
For Enter a value, choose Enter custom value and enter RESELLER NAME.
For Flag result value as, choose Custom value.
For Value if true, choose Select source column and set Value of to Column_2.
For Value if false, choose Enter custom value and enter INVALID.
Choose Apply.

Your dataset should now look like the following screenshot, with the Reseller Name value extracted to a column by itself.

Next, you remove invalid rows and fill the rows with the Reseller Name value.

Choose Clean and then choose Custom values.
For Source column, choose ResellerName.
For Specify values to remove, choose Custom value.
For Values to remove, choose Invalid.
For Apply transform to, choose All rows.
Choose Apply.
Choose Missing and then choose Fill with most frequent value.
For Source column, choose FirstTransactionDate.
For Missing value action, choose Fill with most frequent value.
For Apply transform to, choose All rows.
Choose Apply.

Your dataset should now look like the following screenshot, with the Reseller Name value extracted to a column by itself.

Repeat the same steps in this section for the rest of the metadata, including Reseller Email Address, Reseller ID, and First Transaction Date.

Promote column headers and clean up data

To promote column headers, complete the following steps:

Reorder the columns to put the metadata columns to the left of the dataset by choosing Column, Move column, and Start of the table.
Rename the columns with the appropriate names.

Now you can clean up some columns and rows.

Delete unnecessary columns, such as Column_7.

You can also delete invalid rows by filtering out records that don’t have a transaction date value.

Choose the ABC icon on the menu of the Transaction_Date column and choose date.
For Handle invalid values, select Delete rows, then choose Apply.

The dataset should now have the metadata extracted and the column headers promoted.

Combine multi-line rows into single-line rows

The next issue to address is transactions pertaining to the same row that are split across multiple lines. In the following steps, we extract the needed data from the rows and merge it into single-line transactions. For this example specifically, the Reseller Margin data is split across two lines.

Complete the following steps to get the Reseller Margin value on the same line as the corresponding transaction. First, we identify the Reseller Margin rows and store them in a temporary column.

Choose Conditions and then choose IF.
For Matching conditions, choose Match all conditions.
For Source, choose Value of and Transaction_ID.
For Logical condition, choose Contains.
For Enter a value, choose Enter custom value and enter Reseller Margin.
For Flag result value as, choose Custom value.
For Value if true, choose Select source column set Value of to TransactionAmount.
For Value if false, choose Enter custom value and enter Invalid.
For Destination column, choose ResellerMargin_Temp.
Choose Apply.

Next, you shift the Reseller Margin value up one row.

Choose Functions and then choose NEXT.
For Source column, choose ResellerMargin_Temp.
For Number of rows, enter 1.
For Destination column, choose ResellerMargin.
For Apply transform to, choose All rows.
Choose Apply.

Next, delete the invalid rows.

Choose Missing and then choose Remove missing rows.
For Source column, choose TransactionDate.
For Missing value action, choose Delete rows with missing values.
For Apply transform to, choose All rows.
Choose Apply.

Your dataset should now look like the following screenshot, with the Reseller Margin value extracted to a column by itself.

With the data structured properly, we can move on to mining the cleaned data.

Extract unique identifiers from within strings and text

Many types of data contain important information stored as unstructured text in a cell. In this section, we look at how to extract this data. Within the sample dataset, the BankTransferText column has valuable information around our resellers’ registered bank account numbers as well as the currency of the transaction, namely IBAN, SWIFT Code, and Currency.

Complete the following steps to extract IBAN, SWIFT code, and Currency into separate columns. First, you extract the IBAN number from the text using a regular expression (regex).

Choose Extract and then choose Custom value or pattern.
For Create column options, choose Extract values.
For Source column, choose BankTransferText.
For Extract options, choose Custom value or pattern.
For Values to extract, enter [a-zA-Z][a-zA-Z][0-9]{2}[A-Z0-9]{1,30}.
For Destination column, choose IBAN.
For Apply transform to, choose All rows.
Choose Apply.
Extract the SWIFT code from the text using a regex following the same steps used to extract the IBAN number, but using the following regex instead: (?!^)(SWIFT Code: )([A-Z]{2}[A-Z0-9]+).

Next, remove the SWIFT Code: label from the extracted text.

Choose Remove and then choose Custom values.
For Source column, choose SWIFT Code.
For Specify values to remove, choose Custom value.
For Apply transform to, choose All rows.
Extract the currency from the text using a regex following the same steps used to extract the IBAN number, but using the following regex instead: (?!^)(Currency: )([A-Z]{3}).
Remove the Currency: label from the extracted text following the same steps used to remove the SWIFT Code: label.

You can clean up by deleting any unnecessary columns.

Choose Column and then choose Delete.
For Source columns, choose BankTransferText.
Choose Apply.
Repeat for any remaining columns.

Your dataset should now look like the following screenshot, with IBAN, SWIFT Code, and Currency extracted to separate columns.

Write the transformed data to Amazon S3

With all the steps captured in the recipe, the last step is to write the transformed data to Amazon S3.

On the DataBrew console, choose Run job.

For Job name, enter FoodMartSalesToDataLake.
For Output to, choose Amazon S3.
For File type, choose CSV.
For Delimiter, choose Comma (,).
For Compression, choose None.
For S3 bucket owners’ account, select Current AWS account.
For S3 location, enter s3://{name of S3 bucket}/clean/.
For Role name, choose the IAM role created as a prerequisite or create a new role.
Choose Create and run job.
Go to the Jobs tab and wait for the job to complete.
Navigate to the job output folder on the Amazon S3 console.
Download the CSV file and view the transformed output.

Your dataset should look similar to the following screenshot.

Clean up

To optimize cost, make sure to clean up the resources deployed for this project by completing the following steps:

Delete every DataBrew project along with their linked recipes.
Delete all the DataBrew datasets.
Delete the contents in your S3 bucket.
Delete the S3 bucket.

Conclusion

The reality of exchanging data with suppliers is that we can’t always control the shape of the input data. With DataBrew, we can use a list of pre-built transformations and repeatable steps to transform incoming data into a desired layout and extract relevant data and insights from Excel or CSV files. Start using DataBrew today and transform 3 ^rd party files into structured datasets ready for consumption by your business.

About the Author

Ismail Makhlouf is a Senior Specialist Solutions Architect for Data Analytics at AWS. Ismail focuses on architecting solutions for organizations across their end-to-end data analytics estate, including batch and real-time streaming, big data, data warehousing, and data lake workloads. He primarily works with direct-to-consumer platform companies in the ecommerce, FinTech, PropTech, and HealthTech space to achieve their business objectives with well-architected data platforms.

Managing AWS Lambda runtime upgrades

2023-11-15 Julian Wood

Post Syndicated from Julian Wood original https://aws.amazon.com/blogs/compute/managing-aws-lambda-runtime-upgrades/

This post is written by Julian Wood, Principal Developer Advocate, and Dan Fox, Principal Specialist Serverless Solutions Architect.

AWS Lambda supports multiple programming languages through the use of runtimes. A Lambda runtime provides a language-specific execution environment, which provides the OS, language support, and additional settings, such as environment variables and certificates that you can access from your function code.

You can use managed runtimes that Lambda provides or build your own. Each major programming language release has a separate managed runtime, with a unique runtime identifier, such as python3.11 or nodejs20.x.

Lambda automatically applies patches and security updates to all managed runtimes and their corresponding container base images. Automatic runtime patching is one of the features customers love most about Lambda. When these patches are no longer available, Lambda ends support for the runtime. Over the next few months, Lambda is deprecating a number of popular runtimes, triggered by end of life of upstream language versions and of Amazon Linux 1.

Runtime	Deprecation
Node.js 14	Nov 27, 2023
Node.js 16	Mar 11, 2024
Python 3.7	Nov 27, 2023
Java 8 (Amazon Linux 1)	Dec 31, 2023
Go 1.x	Dec 31, 2023
Ruby 2.7	Dec 07, 2023
Custom Runtime (provided)	Dec 31, 2023

Runtime deprecation is not unique to Lambda. You must upgrade code using Python 3.7 or Node.js 14 when those language versions reach end of life, regardless of which compute service your code is running on. Lambda can help make this easier by tracking which runtimes you are using and providing deprecation notifications.

This post contains considerations and best practices for managing runtime deprecations and upgrades when using Lambda. Adopting these techniques makes managing runtime upgrades easier, especially when working with a large number of functions.

Specifying Lambda runtimes

When you deploy your function as a .zip file archive, you choose a runtime when you create the function. To change the runtime, you can update your function’s configuration.

Lambda keeps each managed runtime up to date by taking on the operational burden of patching the runtimes with security updates, bug fixes, new features, performance enhancements, and support for minor version releases. These runtime updates are published as runtime versions. Lambda applies runtime updates to functions by migrating the function from an earlier runtime version to a new runtime version.

You can control how your functions receive these updates using runtime management controls. Runtime versions and runtime updates apply to patch updates for a given Lambda runtime. Lambda does not automatically upgrade functions between major language runtime versions, for example, from nodejs14.x to nodejs18.x.

For a function defined as a container image, you choose a runtime and the Linux distribution when you create the container image. Most customers start with one of the Lambda base container images, although you can also build your own images from scratch. To change the runtime, you create a new container image from a different base container image.

Why does Lambda deprecate runtimes?

Lambda deprecates a runtime when upstream runtime language maintainers mark their language end-of-life or security updates are no longer available.

In almost all cases, the end-of-life date of a language version or operating system is published well in advance. The Lambda runtime deprecation policy gives end-of-life schedules for each language that Lambda supports. Lambda notifies you by email and via your Personal Health Dashboard if you are using a runtime that is scheduled for deprecation.

Lambda runtime deprecation happens in several stages. Lambda first blocks creating new functions that use a given runtime. Lambda later also blocks updating existing functions using the unsupported runtime, except to update to a supported runtime. Lambda does not block invocations of functions that use a deprecated runtime. Function invocations continue indefinitely after the runtime reaches end of support.

Lambda is extending the deprecation notification period from 60 days before deprecation to 180 days. Previously, blocking new function creation happened at deprecation and blocking updates to existing functions 30 days later. Blocking creation of new functions now happens 30 days after deprecation, and blocking updates to existing functions 60 days after.

Lambda occasionally delays deprecation of a Lambda runtime for a limited period beyond the end of support date of the language version that the runtime supports. During this period, Lambda only applies security patches to the runtime OS. Lambda doesn’t apply security patches to programming language runtimes after they reach their end of support date.

Can Lambda automatically upgrade my runtime?

Moving from one major version of the language runtime to another has a significant risk of being a breaking change. Some libraries and dependencies within a language have deprecation schedules and do not support versions of a language past a certain point. Moving functions to new runtimes could potentially impact large-scale production workloads that customers depend on.

Since Lambda cannot guarantee backward compatibility between major language versions, upgrading the Lambda runtime used by a function is a customer-driven operation.

Lambda function versions

You can use function versions to manage the deployment of your functions. In Lambda, you make code and configuration changes to the default function version, which is called $LATEST. When you publish a function version, Lambda takes a snapshot of the code, runtime, and function configuration to maintain a consistent experience for users of that function version. When you invoke a function, you can specify the version to use or invoke the $LATEST version. Lambda function versions are required when using Provisioned Concurrency or SnapStart.

Some developers use an auto-versioning process by creating a new function version each time they deploy a change. This results in many versions of a function, with only a single version actually in use.

While Lambda applies runtime updates to published function versions, you cannot update the runtime major version for a published function version, for example from Node.js 16 to Node.js 20. To update the runtime for a function, you must update the $LATEST version, then create a new published function version if necessary. This means that different versions of a function can use different runtimes. The following shows the same function with version 1 using Node.js 14.x and version 2 using Node.js 18.x.

Version 1 using Node.js 14.x

Version 2 using Node.js 18.x

Ensure you create a maintenance process for deleting unused function versions, which also impact your Lambda storage quota.

Managing function runtime upgrades

Managing function runtime upgrades should be part of your software delivery lifecycle, in a similar way to how you treat dependencies and security updates. You need to understand which functions are being actively used in your organization. Organizations can create prioritization based on security profiles and/or function usage. You can use the same communication mechanisms you may already be using for handling security vulnerabilities.

Implement preventative guardrails to ensure that developers can only create functions using supported runtimes. Using infrastructure as code, CI/CD pipelines, and robust testing practices makes updating runtimes easier.

Identifying impacted functions

There are tools available to check Lambda runtime configuration and to identify which functions and what published function versions are actually in use. Deleting a function or function version that is no longer in use is the simplest way to avoid runtime deprecations.

You can identify functions using deprecated or soon to be deprecated runtimes using AWS Trusted Advisor. Use the AWS Lambda Functions Using Deprecated Runtimes check, in the Security category that provides 120 days’ notice.

AWS Trusted Advisor Lambda functions using deprecated runtimes

Trusted Advisor scans all versions of your functions, including $LATEST and published versions.

The AWS Command Line Interface (AWS CLI) can list all functions in a specific Region that are using a specific runtime. To find all functions in your account, repeat the following command for each AWS Region and account. Replace the <REGION> and <RUNTIME> parameters with your values. The --function-version ALL parameter causes all function versions to be returned; omit this parameter to return only the $LATEST version.

aws lambda list-functions --function-version ALL --region <REGION> --output text —query "Functions[?Runtime=='<RUNTIME>'].FunctionArn"

You can use AWS Config to create a view of the configuration of resources in your account and also store configuration snapshot data in Amazon S3. AWS Config queries do not support published function versions, they can only query the $LATEST version.

You can then use Amazon Athena and Amazon QuickSight to make dashboards to visualize AWS Config data. For more information, see the Implementing governance in depth for serverless applications learning guide.

Dashboard showing AWS Config data

There are a number of ways that you can track Lambda function usage.

You can use Amazon CloudWatch metrics explorer to view Lambda by runtime and track the Invocations metric within the default CloudWatch metrics retention period of 15 months.

Track invocations in Amazon CloudWatch metrics

You can turn on AWS CloudTrail data event logging to log an event every time Lambda functions are invoked. This helps you understand what identities are invoking functions and the frequency of their invocations.

AWS Cost and Usage Reports can show which functions are incurring cost and in use.

Limiting runtime usage

AWS CloudFormation Guard is an open-source evaluation tool to validate infrastructure as code templates. Create policy rules to ensure that developers only chose approved runtimes. For more information, see Preventative Controls with AWS CloudFormation Guard.

AWS Config rules allow you to check that Lambda function settings for the runtime match expected values. For more information on running these rules before deployment, see Preventative Controls with AWS Config. You can also reactively flag functions as non-compliant as your governance policies evolve. For more information, see Detective Controls with AWS Config.

Lambda does not currently have service control policies (SCP) to block function creation based on the runtime

Upgrade best practices

Use infrastructure as code tools to build and manage your Lambda functions, which can make it easier to manage upgrades.

Ensure you run tests against your functions when developing locally. Include automated tests as part of your CI/CD pipelines to provide confidence in your runtime upgrades. When rolling out function upgrades, you can use weighted aliases to shift traffic between two function versions as you monitor for errors and failures.

Using runtimes after deprecation

AWS strongly advises you to upgrade your functions to a supported runtime before deprecation to continue to benefit from security patches, bug-fixes, and the latest runtime features. While deprecation does not affect function invocations, you will be using an unsupported runtime, which may have unpatched security vulnerabilities. Your function may eventually stop working, for example, due to a certificate expiry.

Lambda blocks function creation and updates for functions using deprecated runtimes. To create or update functions after these operations are blocked, contact AWS Support.

Conclusion

Lambda is deprecating a number of popular runtimes over the next few months, reflecting the end-of-life of upstream language versions and Amazon Linux 1. This post covers considerations for managing Lambda function runtime upgrades.

For more serverless learning resources, visit Serverless Land.

Node.js 20.x runtime now available in AWS Lambda

2023-11-15 Pascal Vogel

Post Syndicated from Pascal Vogel original https://aws.amazon.com/blogs/compute/node-js-20-x-runtime-now-available-in-aws-lambda/

This post is written by Pascal Vogel, Solutions Architect, and Andrea Amorosi, Senior Solutions Architect.

You can now develop AWS Lambda functions using the Node.js 20 runtime. This Node.js version is in active LTS status and ready for general use. To use this new version, specify a runtime parameter value of nodejs20.x when creating or updating functions or by using the appropriate container base image.

You can use Node.js 20 with Powertools for AWS Lambda (TypeScript), a developer toolkit to implement serverless best practices and increase developer velocity. Powertools for AWS Lambda includes proven libraries to support common patterns such as observability, Parameter Store integration, idempotency, batch processing, and more.

You can also use Node.js 20 with Lambda@Edge, allowing you to customize low-latency content delivered through Amazon CloudFront.

This blog post highlights important changes to the Node.js runtime, notable Node.js language updates, and how you can use the new Node.js 20 runtime in your serverless applications.

Node.js 20 runtime updates

Changes to Root CA certificate loading

By default, Node.js includes root certificate authority (CA) certificates from well-known certificate providers. Earlier Lambda Node.js runtimes up to Node.js 18 augmented these certificates with Amazon-specific CA certificates, making it easier to create functions accessing other AWS services. For example, it included the Amazon RDS certificates necessary for validating the server identity certificate installed on your Amazon RDS database.

However, loading these additional certificates has a performance impact during cold start. Starting with Node.js 20, Lambda no longer loads these additional CA certificates by default. The Node.js 20 runtime contains a certificate file with all Amazon CA certificates located at /var/runtime/ca-cert.pem. By setting the NODE_EXTRA_CA_CERTS environment variable to /var/runtime/ca-cert.pem, you can restore the behavior from Node.js 18 and earlier runtimes.

This causes Node.js to validate and load all Amazon CA certificates during a cold start. It can take longer compared to loading only specific certificates. For the best performance, we recommend bundling only the certificates that you need with your deployment package and loading them via NODE_EXTRA_CA_CERTS. The certificates file should consist of one or more trusted root or intermediate CA certificates in PEM format.

For example, for RDS, include the required certificates alongside your code as certificates/rds.pem and then load it as follows:

NODE_EXTRA_CA_CERTS=/var/task/certificates/rds.pem

See Using Lambda environment variables in the AWS Lambda Developer Guide for detailed instructions for setting environment variables.

Amazon Linux 2023

The Node.js 20 runtime is based on the provided.al2023 runtime. The provided.al2023 runtime in turn is based on the Amazon Linux 2023 minimal container image release and brings several improvements over Amazon Linux 2 (AL2).

provided.al2023 contains only the essential components necessary to install other packages and offers a smaller deployment footprint with a compressed image size of less than 40MB compared to the over 100MB AL2-based base image.

With glibc version 2.34, customers have access to a more recent version of glibc, updated from version 2.26 in AL2-based images.

The Amazon Linux 2023 minimal image uses microdnf as package manager, symlinked as dnf, replacing yum in AL2-based images. Additionally, curl and gnupg2 are also included as their minimal versions curl-minimal and gnupg2-minimal.

Learn more about the provided.al2023 runtime in the blog post Introducing the Amazon Linux 2023 runtime for AWS Lambda and the Amazon Linux 2023 launch blog post.

Runtime Interface Client

The Node.js 20 runtime uses the open source AWS Lambda NodeJS Runtime Interface Client (RIC). You can now use the same RIC version in your Open Container Initiative (OCI) Lambda container images as the one used by the managed Node.js 20 runtime.

The Node.js 20 runtime supports Lambda response streaming which enables you to send response payload data to callers as it becomes available. Response streaming can improve application performance by reducing time-to-first-byte, can indicate progress during long-running tasks, and allows you to build functions that return payloads larger than 6MB, which is the Lambda limit for buffered responses.

Setting Node.js heap memory size

Node.js allows you to configure the heap size of the v8 engine via the --max-old-space-size and --max-semi-space-size options. By default, Lambda overrides the Node.js default values with values derived from the memory size configured for the function. If you need control over your runtime’s memory allocation, you can now set both of these options using the NODE_OPTIONS environment variable, without needing an exec wrapper script. See Using Lambda environment variables in the AWS Lambda Developer Guide for details.

Use the --max-old-space-size option to set the max memory size of V8’s old memory section, and the --max-semi-space-size option to set the maximum semispace size for V8’s garbage collector. See the Node.js documentation for more details on these options.

Node.js 20 language updates

Language features

With this release, Lambda customers can take advantage of new Node.js 20 language features, including:

HTTP(S)/1.1 default keepAlive: Node.js now sets keepAlive to true by default. Any outgoing HTTPs connections use HTTP 1.1 keep-alive with a default waiting window of 5 seconds. This can deliver improved throughput as connections are reused by default.
Fetch API is enabled by default: The global Node.js Fetch API is enabled by default. However, it is still an experimental module.
Faster URL parsing: Node.js 20 comes with the Ada 2.0 URL parser which brings performance improvements to URL parsing. This has also been back-ported to Node.js 18.7.0.
Web Crypto API now stable: The Node.js implementation of the standard Web Crypto API has been marked as stable. You can access the provided cryptographic primitives through globalThis.crypto.
Web assembly support: Node.js 20 enables the experimental WebAssembly System Interface (WASI) API by default without the need to set an experimental flag.

For a detailed overview of Node.js 20 language features, see the Node.js 20 release blog post and the Node.js 20 changelog.

Performance considerations

Node.js 19.3 introduced a change that impacts how non-essential modules are lazy-loaded during the Node.js process startup. In terms of the impact to your Lambda functions, this reduces the work during initialization of each execution environment, then if used, the modules will instead be loaded during the first function invoke. This change remains in Node.js 20.

Builders should continue to measure and test function performance and optimize function code and configuration for any impact. To learn more about how to optimize Node.js performance in Lambda, see Performance optimization in the Lambda Operator Guide, and our blog post Optimizing Node.js dependencies in AWS Lambda.

Migration from earlier Node.js runtimes

Migration from Node.js 16

In the case of Node.js 16, we have delayed deprecation from the community end of support date on September 11, 2023, to June 12, 2024. This gives customers the opportunity to migrate directly from Node.js 16 to Node.js 20, skipping Node.js 18.

AWS SDK for JavaScript

Up until Node.js 16, Lambda’s Node.js runtimes included the AWS SDK for JavaScript version 2. This has since been superseded by the AWS SDK for JavaScript version 3, which was released in December 2020. Starting with Node.js 18, and continuing with Node.js 20, the Lambda Node.js runtimes have upgraded the version of the AWS SDK for JavaScript included in the runtime from v2 to v3. Customers upgrading from Node.js 16 or earlier runtimes who are using the included AWS SDK for JavaScript v2 should upgrade their code to use the v3 SDK.

For optimal performance, and to have full control over your code dependencies, we recommend bundling and minifying the AWS SDK in your deployment package, rather than using the SDK included in the runtime. For more information, see Optimizing Node.js dependencies in AWS Lambda.

Using the Node.js 20 runtime in AWS Lambda

AWS Management Console

To use the Node.js 20 runtime to develop your Lambda functions, specify a runtime parameter value Node.js 20.x when creating or updating a function. The Node.js 20 runtime version is now available in the Runtime dropdown on the Create function page in the AWS Lambda console:

To update an existing Lambda function to Node.js 20, navigate to the function in the Lambda console, then choose Edit in the Runtime settings panel. The new version of Node.js is available in the Runtime dropdown:

AWS Lambda – Container Image

Change the Node.js base image version by modifying the FROM statement in your Dockerfile:


FROM public.ecr.aws/lambda/nodejs:20
# Copy function code
COPY lambda_handler.xx ${LAMBDA_TASK_ROOT}

Customers running Node.js 20 Docker images locally, including customers using AWS SAM, will need to upgrade their Docker install to version 20.10.10 or later.

AWS Serverless Application Model (AWS SAM)

In AWS SAM, set the Runtime attribute to node20.x to use this version:


AWSTemplateFormatVersion: "2010-09-09"
Transform: AWS::Serverless-2016-10-31

Resources:
  MyFunction:
    Type: AWS::Serverless::Function
    Properties:
      Handler: lambda_function.lambda_handler
      Runtime: nodejs20.x
      CodeUri: my_function/.
      Description: My Node.js Lambda Function

AWS Cloud Development Kit (AWS CDK)

In the AWS CDK, set the runtime attribute to Runtime.NODEJS_20_X to use this version:


import * as cdk from "aws-cdk-lib";
import * as lambda from "aws-cdk-lib/aws-lambda";
import * as path from "path";
import { Construct } from "constructs";

export class CdkStack extends cdk.Stack {
  constructor(scope: Construct, id: string, props?: cdk.StackProps) {
    super(scope, id, props);

    // The code that defines your stack goes here

    // The Node.js 20 enabled Lambda Function
    const lambdaFunction = new lambda.Function(this, "node20LambdaFunction", {
      runtime: lambda.Runtime.NODEJS_20_X,
      code: lambda.Code.fromAsset(path.join(__dirname, "/../lambda")),
      handler: "index.handler",
    });
  }
}

Conclusion

Lambda now supports Node.js 20. This release uses the Amazon Linux 2023 OS, supports configurable CA certificate loading for faster cold starts, as well as other improvements detailed in this blog post.

You can build and deploy functions using the Node.js 20 runtime using the AWS Management Console, AWS CLI, AWS SDK, AWS SAM, AWS CDK, or your choice of Infrastructure as Code (IaC). You can also use the Node.js 20 container base image if you prefer to build and deploy your functions using container images.

The Node.js 20 runtime empowers developers to build more efficient, powerful, and scalable serverless applications. Try the Node.js runtime in Lambda today and read about the Node.js programming model in the Lambda documentation to learn more about writing functions in Node.js 20.

For more serverless learning resources, visit Serverless Land.

Converting Apache Kafka events from Avro to JSON using EventBridge Pipes

2023-11-15 Pascal Vogel

Post Syndicated from Pascal Vogel original https://aws.amazon.com/blogs/compute/converting-apache-kafka-events-from-avro-to-json-using-eventbridge-pipes/

This post is written by Pascal Vogel, Solutions Architect, and Philipp Klose, Global Solutions Architect.

Event streaming with Apache Kafka has become an important element of modern data-oriented and event-driven architectures (EDAs), unlocking use cases such as real-time analytics of user behavior, anomaly and fraud detection, and Internet of Things event processing. Stream producers and consumers in Kafka often use schema registries to ensure that all components follow agreed-upon event structures when sending (serializing) and processing (deserializing) events to avoid application bugs and crashes.

A common schema format in Kafka is Apache Avro, which supports rich data structures in a compact binary format. To integrate Kafka with other AWS and third-party services more easily, AWS offers Amazon EventBridge Pipes, a serverless point-to-point integration service. However, many downstream services expect JSON-encoded events, requiring custom, and repetitive schema validation and conversion logic from Avro to JSON in each downstream service.

This blog post shows how to reliably consume, validate, convert, and send Avro events from Kafka to AWS and third-party services using EventBridge Pipes, allowing you to reduce custom deserialization logic in downstream services. You can also use EventBridge event buses as targets in Pipes to filter and distribute events from Pipes to multiple targets, including cross-account and cross-Region delivery.

This blog describes two scenarios:

Using Amazon Managed Streaming for Apache Kafka (Amazon MSK) and AWS Glue Schema Registry.
Using Confluent Cloud and the Confluent Schema Registry.

See the associated GitHub repositories for Glue Schema Registry or Confluent Schema Registry for full source code and detailed deployment instructions.

Kafka event streaming and schema validation on AWS

To build event streaming applications with Kafka on AWS, you can use Amazon MSK, offerings such as Confluent Cloud, or self-hosted Kafka on Amazon Elastic Compute Cloud (Amazon EC2) instances.

To avoid common issues in event streaming and event-driven architectures, such as data inconsistencies and incompatibilities, it is a recommended practice to define and share event schemas between event producers and consumers. In Kafka, schema registries are used to manage, evolve, and enforce schemas for event producers and consumers. The AWS Glue Schema Registry provides a central location to discover, manage, and evolve schemas. In the case of Confluent Cloud, the Confluent Schema Registry serves the same role. Both the Glue Schema Registry and the Confluent Schema Registry support common schema formats such as Avro, Protobuf, and JSON.

To integrate Kafka with AWS services, third-party services, and your own applications, you can use EventBridge Pipes. EventBridge Pipes helps you create point-to-point integrations between event sources and targets with optional filtering, transformation, and enrichment. EventBridge Pipes reduces the amount of integration code that you have to write and maintain when building EDAs.

Many AWS and third-party services expect JSON-encoded payloads (events) as input, meaning they cannot directly consume Avro or Protobuf payloads. To replace repetitive Avro-to-JSON validation and conversion logic in each consumer, you can use the EventBridge Pipes enrichment step. This solution uses an AWS Lambda function in the enrichment step to deserialize and validate Kafka events with a schema registry, including error handling with dead-letter queues, and convert events to JSON before passing them to downstream services.

Solution overview

The solution presented in this blog post consists of the following key elements:

The source of the pipe is a Kafka cluster deployed using MSK or Confluent Cloud. EventBridge Pipes reads events from the Kafka stream in batches and sends them to the enrichment function (see here for an example event).
The enrichment step (Lambda function) deserializes and validates the events against the configured schema registry (Glue or Confluent), converts events from Avro to JSON with integrated error handling, and returns them to the pipe.
The target of this example solution is an EventBridge custom event bus that is invoked by EventBridge Pipes with JSON-encoded events returned by the enrichment Lambda function. EventBridge Pipes supports a variety of other targets, including Lambda, AWS Step Functions, Amazon API Gateway, API destinations, and more, enabling you to build EDAs without writing integration code.
In this sample solution, the event bus sends all events to Amazon CloudWatch Logs via an EventBridge rule. You can extend the example to invoke additional EventBridge targets.

Optionally, you can add OpenAPI 3 or JSONSchema Draft 4 schemas for your events in the EventBridge schema registry by either manually generating it from the Avro schema or using EventBridge schema discovery. This allows you to download code bindings for the JSON-converted events for various programming languages, such as JavaScript, Python, and Java, to correctly use them in your EventBridge targets.

The remainder of this blog post describes this solution for the Glue and Confluent schema registries with code examples.

EventBridge Pipes with the Glue Schema Registry

This section describes how to implement event schema validation and conversion from Avro to JSON using EventBridge Pipes and the Glue Schema Registry. You can find the source code and detailed deployment instructions on GitHub.

Prerequisites

You need an Amazon MSK serverless cluster running and the Glue Schema registry configured. This example includes a Avro schema and a Glue Schema Registry. See the following AWS blog post for an introduction to schema validation with the Glue Schema Registry: Validate, evolve, and control schemas in Amazon MSK and Amazon Kinesis Data Streams with AWS Glue Schema Registry.

EventBridge Pipes configuration

Use the AWS Cloud Development Kit (AWS CDK) template provided in the GitHub repository to deploy:

An EventBridge pipe that connects to your existing Amazon MSK Serverless Kafka topic as the source via AWS Identity and Access Management (IAM) authentication.
EventBridge Pipes reads events from your Kafka topic using the Amazon MSK source type.
An enrichment Lambda function in Java to perform event deserialization, validation, and conversion from Avro to JSON.
An Amazon Simple Queue Service (Amazon SQS) dead letter queue to hold events for which deserialization failed.
An EventBridge custom event bus as the pipe target. An EventBridge rule sends all incoming events into a CloudWatch Logs log group.

For MSK-based sources, EventBridge supports configuration parameters, such as batch window, batch size, and starting position, which you can set using the parameters of the CfnPipe class in the example CDK stack.

The example EventBridge pipe consumes events from Kafka in batches of 10 because it is targeting an EventBridge event bus, which has a max batch size of 10. See batching and concurrency in the EventBridge Pipes User Guide to choose an optimal configuration for other targets.

EventBridge Pipes with the Confluent Schema Registry

This section describes how to implement event schema validation and conversion from Avro to JSON using EventBridge Pipes and the Confluent Schema Registry. You can find the source code and detailed deployment instructions on GitHub.

Prerequisites

To set up this solution, you need a Kafka stream running on Confluent Cloud as well as the Confluent Schema Registry set up. See the corresponding Schema Registry tutorial for Confluent Cloud to set up a schema registry for your Confluent Kafka stream.

To connect to your Confluent Cloud Kafka cluster, you need an API key for Confluent Cloud and Confluent Schema Registry. AWS Secrets Manager is used to securely store your Confluent secrets.

EventBridge Pipes configuration

Use the AWS CDK template provided in the GitHub repository to deploy:

An EventBridge pipe that connects to your existing Confluent Kafka topic as the source via an API secret stored in Secrets Manager.
EventBridge Pipes reads events from your Confluent Kafka topic using the self-managed Apache Kafka stream source type, which includes all non-MSK Kafka clusters.
An enrichment Lambda function in Python to perform event deserialization, validation, and conversion from Avro to JSON.
An SQS dead letter queue to hold events for which deserialization failed.
An EventBridge custom event bus as the pipe target. An EventBridge rule writes all incoming events into a CloudWatch Logs log group.

For self-managed Kafka sources, EventBridge supports configuration parameters, such as batch window, batch size, and starting position, which you can set using the parameters of the CfnPipe class in the example CDK stack.

Enrichment Lambda functions

Both of the solutions described previously include an enrichment Lambda function for schema validation and conversion from Avro to JSON.

The Java Lambda function integrates with the Glue Schema Registry using the AWS Glue Schema Registry Library. The Python Lambda function integrates with the Confluent Schema Registry using the confluent-kafka library and uses Powertools for AWS Lambda (Python) to implement Serverless best practices such as logging and tracing.

The enrichment Lambda functions perform the following tasks:

In the events polled from the Kafka stream by the EventBridge pipe, the key and value of the event are base64 encoded. Therefore, for each event in the batch passed to the function, the key and the value are decoded.
The event key is assumed to be serialized by the producer as a string type.
The event value is deserialized using the Glue Schema registry Serde (Java) or the confluent-kafka AvroDeserializer (Python).
The function then returns the successfully converted JSON events to the EventBridge pipe, which then invokes the target for each of them.
Events for which Avro deserialization failed are sent to the SQS dead letter queue.

Conclusion

This blog post shows how to implement event consumption, Avro schema validation, and conversion to JSON using Amazon EventBridge Pipes, Glue Schema Registry, and Confluent Schema Registry.

The source code for the presented example is available in the AWS Samples GitHub repository for Glue Schema Registry and Confluent Schema Registry. For more patterns, visit the Serverless Patterns Collection.

For more serverless learning resources, visit Serverless Land.

Introducing logging support for Amazon EventBridge Pipes

2023-11-15 David Boyne

Post Syndicated from David Boyne original https://aws.amazon.com/blogs/compute/introducing-logging-support-for-amazon-eventbridge-pipes/

Today, AWS is announcing support for logging with EventBridge Pipes. Amazon EventBridge Pipes is a point to point integration solution that connects event producers and consumers with optional filter, transform, and enrichment steps. EventBridge Pipes reduces the amount of integration code builders must write and maintain when building event-driven applications. Popular integrations include connecting Amazon Kinesis streams together with filtering, Amazon DynamoDB direct integrations with Amazon EventBridge, and Amazon SQS integrations with AWS Step Functions.

Amazon EventBridge Pipes

EventBridge Pipes logging introduces insights into different stages of the pipe execution. It expands on the Amazon CloudWatch metrics support, and provides you with additional methods for troubleshooting and debugging.

You can now gain insights into various successful and failure scenarios within the pipe execution steps. When event transformation or enrichment succeeds or fails, you can use logs to delve deeper and initiate troubleshooting for any issues with their configured pipes.

EventBridge Pipes execution steps

Understanding pipes execution steps can assist you in selecting the appropriate log level, which determines how much information is logged.

A pipe execution is an event or batch of events received by a pipe that travel from the source to the target. As events travel through a pipe, they can be filtered, transformed, or enriched using AWS Step Functions, AWS Lambda, Amazon API Gateway, and EventBridge API Destinations.

A pipe execution consists of two main stages: the enrichment and target. Both of these stages encompass transformation and invocation steps.

You can use input transformers, enabling modification to the payload of the event before the events undergo enrichment or get dispatched to the downstream target. This gives you fine-grained control over the manipulation of event data during the execution of their configured pipes.

When the pipe execution starts, the execution enters the enrichment stage. If you don’t configure an enrichment stage, the execution proceeds to the target stage.

At the pipe execution, transformation, enrichment, and target phases EventBridge can log information to help debug or troubleshoot. Pipes logs can include payloads, errors, transformations, AWS requests, and AWS responses.

To learn more about pipe executions, read this documentation.

Configuring log levels with EventBridge Pipes

When logging is enabled for your pipe, EventBridge produces a log entry for every execution step and sends these logs to the specified log destinations.

EventBridge Pipes supports three log destinations: Amazon CloudWatch Logs, Amazon Kinesis Data Firehose stream, and Amazon S3. The records sent can be customized by configuring the log level of the pipe (OFF, ERROR, INFO, TRACE).

OFF – EventBridge does not send any records.
ERROR – EventBridge sends records related to errors generated during the pipe execution. Examples include Execution Failed, Execution Timeout and Enrichment Failures.
INFO – EventBridge sends records related to errors and selected information performed during pipe execution. Examples include Execution Started, Execution Succeeded and Enrichment Stage Succeeded.
TRACE – EventBridge sends any record generated during any step in the pipe execution.

The ERROR log level proves beneficial in gaining insights into the reasons behind a failed pipe execution. Pipe executions may encounter failure due to various reasons, such as timeouts, enrichment failure, transformation failure, or target invocation failure. Enabling ERROR logging allows you to learn more about the specific cause of the pipe error, facilitating the resolution of the issue.

The INFO log level supplements ERROR information with additional details. It not only informs about errors but also provides insights into the commencement of the pipe execution, entry into the enrichment phase, progression into the transformation phase, and the initiation and successful completion of the target stage.

For a more in-depth analysis, you can use the TRACE log level to obtain comprehensive insights into a pipe execution. This encompasses all supported pipe logs, offering a detailed view beyond the INFO and ERROR logs. The TRACE log level reveals crucial information, such as skipped pipe execution stages and the initiation of transformation and enrichment processes.

For more details on the log levels and what logs are sent, you can read the documentation.

Including execution data with EventBridge Pipes logging

To help with further debugging, you can choose to include execution data within the pipe logs. This data comprises event payloads, AWS requests, and responses sent to and received from configured enrichment and target components.

You can also use the execution data to gain further insights into the payloads, requests, and responses sent to AWS services during the execution of the pipe.

Incorporating execution data can enhance understanding of the pipe execution, providing deeper insights and aiding in the debugging of any encountered issues.

Execution data within a log contains three parts:

payload: The content of the event itself. The payload of the event may contain sensitive information and EventBridge makes no attempt to redact the contents. Including execution data is optional and can be turned off.
awsRequest: The request sent to the enrichment or target in serialized JSON format. For API destinations this includes the HTTP request sent to that endpoint.
awsResponse: The response returned by the enrichment or target in JSON format. For API Destinations, this is the response returned from the configured endpoint.

The payload of the event is populated when the event itself can be updated. These stages include the initial pipe execution, enrichment phase, and target phase. The awsRequest and awsResponse are both generated at the final steps of enrichment and targeting.

For more information on log levels and execution data, visit this documentation.

Getting started with EventBridge Pipes logs

This example creates a pipe with logging enabled and includes execution data. The pipe connects two Amazon SQS queues using an input transformer on the target with no enrichment step. The input transformer customizes the payload of the event before reaching the target.

Creating the source and target queues

# Create a queue for the source
aws sqs create-queue --queue-name pipe-source

# Create a queue for the target
aws sqs create-queue --queue-name pipe-target

2. Navigate to EventBridge Pipes and choose Create pipe.

3. Select SQS as the Source and select pipe-source as the SQS Queue.

4. Skip the filter and enrichment phase and add a new Target. Select SQS as the Target service and pipe-target as the Queue.

5. Open Target Input Transformer section and enter the transformer code into the Transformer field.

{

  "body": "Favorite food is <$.body>"

}

6. Choose Pipe settings to configure the log group for the new pipe.

7. Verify that CloudWatch Logs is set as the log destination, and select Trace as the log level. Check the “Include execution data” check box. This logs all traces to the new CloudWatch log group and includes the SQS messages that are sent on the pipe.

8. Choose Create Pipe.

9. Send an SQS message to the source queue.

# Get the Queue URL

aws sqs get-queue-url --queue-name pipe-source

# Send a message to the queue using the URL

aws sqs send-message --queue-url {QUEUE_URL} --message-body "pizza"

10. All trace logs are shown in the monitoring tab, use CloudWatch Logs section for more information.

Conclusion

EventBridge Pipes enables point-to-point integration between event producers and consumers. With logging support for EventBridge Pipes, you can now gain insights into various stages of the pipe execution. Pipe log destinations can be configured to CloudWatch Logs, Kinesis Data Firehose, and Amazon S3.

EventBridge Pipes supports three log levels. The ERROR log level configures EventBridge to send records related to errors to the log destination. The INFO log level configures EventBridge to send records related to errors and selected information during the pipe execution. The TRACE log level sends any record generated to the log destination, useful for debugging and gaining further insights.

You can include execution data in the logs, which includes the event itself, and AWS requests and responses made to AWS services configured in the pipe. This can help you gain further insights into the pipe execution. Read the documentation to learn more about EventBridge Pipes Logs.

For more serverless learning resources, visit Serverless Land.

How Wallapop improved performance of analytics workloads with Amazon Redshift Serverless and data sharing

2023-11-14 Eduard Lopez

Post Syndicated from Eduard Lopez original https://aws.amazon.com/blogs/big-data/how-wallapop-improved-performance-of-analytics-workloads-with-amazon-redshift-serverless-and-data-sharing/

Amazon Redshift is a fast, fully managed cloud data warehouse that makes it straightforward and cost-effective to analyze all your data at petabyte scale, using standard SQL and your existing business intelligence (BI) tools. Today, tens of thousands of customers run business-critical workloads on Amazon Redshift.

Amazon Redshift Serverless makes it effortless to run and scale analytics workloads without having to manage any data warehouse infrastructure.

Redshift Serverless automatically provisions and intelligently scales data warehouse capacity to deliver fast performance for even the most demanding and unpredictable workloads, and you pay only for what you use.

This is ideal when it’s difficult to predict compute needs such as variable workloads, periodic workloads with idle time, and steady-state workloads with spikes. As your demand evolves with new workloads and more concurrent users, Redshift Serverless automatically provisions the right compute resources, and your data warehouse scales seamlessly and automatically.

Amazon Redshift data sharing allows you to securely share live, transactionally consistent data in one Redshift data warehouse with another Redshift data warehouse (provisioned or serverless) across accounts and Regions without needing to copy, replicate, or move data from one data warehouse to another.

Amazon Redshift data sharing enables you to evolve your Amazon Redshift deployment architectures into a hub-and-spoke or data mesh model to better meet performance SLAs, provide workload isolation, perform cross-group analytics, and onboard new use cases, all without the complexity of data movement and data copies.

In this post, we show how Wallapop adopted Redshift Serverless and data sharing to modernize their data warehouse architecture.

Wallapop’s initial data architecture platform

Wallapop is a Spanish ecommerce marketplace company focused on second-hand items, founded in 2013. Every day, they receive around 300,000 new items from buyers to be added to their catalog. The marketplace can be accessed via mobile app or website.

The average monthly traffic is around 15 million active users. Since its creation in 2013, it has reached more than 40 million downloads and more than 700 million products have been listed.

Amazon Redshift plays a central role in their data platform on AWS for ingestion, ETL (extract, transform, and load), machine learning (ML), and consumption workloads that run their insight consumption to drive decision-making.

The initial architecture is composed of one main Redshift provisioned cluster that handled all the workloads, as illustrated in the following diagram. Their cluster was deployed with 8 nodes ra3.4xlarge and concurrency scaling enabled.

Wallapop had three main areas to improve in their initial data architecture platform:

Workload isolation challenges with growing data volumes and new workloads running in parallel
Administrative burden on data engineering teams to manage the concurrent workloads, especially at peak times
Cost-performance ratio while scaling during peak periods

The areas of improvement mainly focused on performance of data consumption workloads along with the BI and analytics consumption tool, where high query concurrency was impacting the final analytics preparation and its insights consumption.

Solution overview

To improve their data platform architecture, Wallapop designed and built a new distributed approach with Amazon Redshift with the support of AWS.

Their cluster size of the provisioned data warehouse didn’t change. What changed was lowering the usage concurrency scaling to 1 hour, which is in the Free Tier usage for every 24 hours of using the main cluster. The following diagram illustrates the target architecture.

Solution details

The new data platform architecture combines Redshift Serverless and provisioned data warehouses with Amazon Redshift data sharing, helping Wallapop improve their overall Amazon Redshift experience with improved ease of use, performance, and optimized costs.

Redshift Serverless measures data warehouse capacity in Redshift Processing Units (RPUs). RPUs are resources used to handle workloads. You can adjust the base capacity setting from 8 RPUs to 512 RPUs in units of 8 (8, 16, 24, and so on).

The new architecture uses a Redshift provisioned cluster with RA3 nodes to run their constant and write workloads (data ingestion and transformation jobs). For cost-efficiency, Wallapop is also benefiting from Redshift reserved instances to optimize on costs for these known, predictable, and steady workloads. This cluster acts as the producer cluster in their distributed architecture using data sharing, meaning the data is ingested into the storage layer of Amazon Redshift—Redshift Managed Storage (RMS).

For the consumption part of the data platform architecture, the data is shared with different Redshift Serverless endpoints to meet the needs for different consumption workloads.

Data sharing provides workloads isolation. With this architecture, Wallapop achieves better workload isolation and ensures that only the right data is shared with the different consumption applications. Additionally, this approach avoids data duplication in their consumer part, which optimizes costs and allows better governance processes, because they only have to manage a single version of the data warehouse data instead of different copies or versions of it.

Redshift Serverless is used as a consumer part of the data platform architecture to meet those predictable and unpredictable, non-steady, and often demanding analytics workloads, such as their CI/CD jobs and BI and analytics consumption workloads coming from their data visualization application. Redshift Serverless also helps them achieve better workload isolation due to its managed auto scaling feature that makes sure performance is consistently good for these unpredictable workloads, even at peak times. It also provides a better user experience for the Wallapop data platform team, thanks to the autonomics capabilities that Redshift Serverless provides.

The new solution combining Redshift Serverless and data sharing allowed Wallapop to achieve better performance, cost, and ease of use.

Eduard Lopez, Wallapop Data Engineering Manager, shared the improved experience of analytics users: “Our analyst users are telling us that now ‘Looker flies.’ Insights consumption went up as a result of it without increasing costs.”

Evaluation of outcome

Wallapop started this re-architecture effort by first testing the isolation of their BI consumption workload with Amazon Redshift data sharing and Redshift Serverless with the support of AWS. The workload was tested using different base RPU configurations to measure the base capacity and resources in Redshift Serverless. Base RPU ranges for Redshift Serverless range from 8–512. Wallapop tested their BI workload with two configurations: 32 base RPU and 64 base RPU, after enabling data sharing from their Redshift provisioned cluster to ensure the serverless endpoints have access to the necessary datasets.

Based on the results measured 1 week before testing, the main area for improvement was the queries that took longer than 10 seconds to complete (52%), represented by the yellow, orange, and red areas of the following chart, as well as the long-running queries represented by the red area (over 600 seconds, 9%).

The first test of this workload with Redshift Serverless using a 64 base RPU configuration immediately showed performance improvement results: the queries running longer than 10 seconds were reduced by 38% and the long-running queries (over 120 seconds) were almost completely eliminated.

Javier Carbajo, Wallapop Data Engineer, says, “Providing a service without downtime or loading times that are too long was one of our main requirements since we couldn’t have analysts or stakeholders without being able to consult the data.”

Following the first set of results, Wallapop also tested with a Redshift Serverless configuration using 32 base RPU to compare the results and select the configuration that could offer them the best price-performance for this workload. With this configuration, the results were similar to the previously test run on Redshift Serverless with 64 base RPU (still showing significant performance improvement from the original results). Based on the tests, this configuration was selected for the new architecture.

Gergely Kajtár, Wallapop Data Engineer, says, “We noticed a significant increase in the daily workflows’ stability after the change to the new Redshift architecture.”

Following this first workload, Wallapop has continued expanding their Amazon Redshift distributed architecture with CI/CD workloads running on a separated Redshift Serverless endpoint using data sharing with their Redshift provisioned (RA3) cluster.

“With the new Redshift architecture, we have noticed remarkable improvements both in speed and stability. That has translated into an increase of 2 times in analytical queries, not only by analysts and data scientists but from other roles as well such as marketing, engineering, C-level, etc. That proves that investing in a scalable architecture like Redshift Serverless has a direct consequence on accelerating the adoption of data as decision-making driver in the organization.”

– Nicolás Herrero, Wallapop Director of Data & Analytics.

Conclusion

In this post, we showed you how this platform can help Wallapop to scale in the future by adding new consumers when new needs or applications require to access data.

If you’re new to Amazon Redshift, you can explore demos, other customer stories, and the latest features at Amazon Redshift. If you’re already using Amazon Redshift, reach out to your AWS account team for support, and learn more about what’s new with Amazon Redshift.

About the Authors

Eduard Lopez is the Data Engineer Manager at Wallapop. He is a software engineer with over 6 years of experience in data engineering, machine learning, and data science.

Daniel Martinez is a Solutions Architect in Iberia Digital Native Businesses (DNB), part of the worldwide commercial sales organization (WWCS) at AWS.

Jordi Montoliu is a Sr. Redshift Specialist in EMEA, part of the worldwide specialist organization (WWSO) at AWS.

Ziad Wali is an Acceleration Lab Solutions Architect at Amazon Web Services. He has over 10 years of experience in databases and data warehousing, where he enjoys building reliable, scalable, and efficient solutions. Outside of work, he enjoys sports and spending time in nature.

Semir Naffati is a Sr. Redshift Specialist Solutions Architect in EMEA, part of the worldwide specialist organization (WWSO) at AWS.

The serverless attendee’s guide to AWS re:Invent 2023

2023-11-14 Marcia Villalba

Post Syndicated from Marcia Villalba original https://aws.amazon.com/blogs/compute/the-serverless-attendees-guide-to-aws-reinvent-2023/

AWS re:Invent 2023 is fast approaching, bringing together tens of thousands of Builders in Las Vegas in November. However, even if you can’t attend in person, you can catch up with sessions on-demand.

Breakout sessions are lecture-style 60-minute informative sessions presented by AWS experts, customers, or partners. These sessions cover beginner (100 level) topics to advanced and expert (300–400 level) topics. The sessions are recorded and uploaded a few days after to the AWS Events YouTube channel.

This post shares the “must watch” breakout sessions related to serverless architectures and services.

Sessions related to serverless architecture

SVS401 | Best practices for serverless developers
Provides architectural best practices, optimizations, and useful shortcuts that experts use to build secure, high-scale, and high-performance serverless applications.

Chris Munns, Startup Tech Leader, AWS
Julian Wood, Principal Developer Advocate, AWS

SVS305 | Refactoring to serverless
Shows how you can refactor your application to serverless with real-life examples.

Gregor Hohpe, Senior Principal Evangelist, AWS
Sindhu Pillai, Senior Solutions Architect, AWS

SVS308 | Building low-latency, event-driven applications
Explores building serverless web applications for low-latency and event-driven support. Marvel Snap share how they achieve low-latency in their games using serverless technology.

Marcia Villalba, Principal Developer Advocate, AWS
Brenna Moore, Second Dinner

SVS309 | Improve productivity by shifting more responsibility to developers
Learn about approaches to accelerate serverless development with faster feedback cycles, exploring best practices and tools. Watch a live demo featuring an improved developer experience for building serverless applications while complying with enterprise governance requirements.

Heeki Park, Principal Solutions Architect, AWS
Sam Dengler, Capital One

GBL203-ES | Building serverless-first applications with MAPFRE
This session is delivered in Spanish. Learn what modern, serverless-first applications are and how to implement them with services such as AWS Lambda or AWS Fargate. Find out how MAPFRE have adopted and implemented a serverless strategy.

Jesus Bernal, Senior Solutions Architect, AWS
Iñigo Lacave, MAPFRE
Mat Jovanovic, MAPFRE

Sessions related to AWS Lambda

BOA311 | Unlocking serverless web applications with AWS Lambda Web Adapter
Learn about the AWS Lambda Web Adapter and how it integrates with familiar frameworks and tools. Find out how to migrate existing web applications to serverless or create new applications using AWS Lambda.

Betty Zheng, Senior Developer Advocate, AWS
Harold Sun, Senior Solutions Architect, AWS

OPN305 | The pragmatic serverless Python developer
Covers an opinionated approach to setting up a serverless Python project, including testing, profiling, deployments, and operations. Learn about many open source tools, including Powertools for AWS Lambda—a toolkit that can help you implement serverless best practices and increase developer velocity.

Heitor Lessa, Principal Solutions Architect, AWS
Ran Isenberg, CyberArk

XNT301 | Build production-ready serverless .NET apps with AWS Lambda
Explores development and architectural best practices when building serverless applications with .NET and AWS Lambda, including when to run ASP.NET on Lambda, code structure, and using native AOT to massively increase performance.

James Eastham, Senior Cloud Architect, AWS
Craig Bossie, Solutions Architect, AWS

COM306 | “Rustifying” serverless: Boost AWS Lambda performance with Rust
Discover how to deploy Rust functions using AWS SAM and cargo-lambda, facilitating a smooth development process from your local machine. Explore how to integrate Rust into Python Lambda functions effortlessly using tools like PyO3 and maturin, along with the AWS SDK for Rust. Uncover how Rust can optimize Lambda functions, including the development of Lambda extensions, all without requiring a complete rewrite of your existing code base.

Efi Merdler-Kravitz, Cloudex

COM305 | Demystifying and mitigating AWS Lambda cold starts
Examines the Lambda initialization process at a low level, using benchmarks comparing common architectural patterns, and then benchmarking various RAM configurations and payload sizes. Next, measure and discuss common mistakes that can increase initialization latency, explore and understand proactive initialization, and learn several strategies you can use to thaw your AWS Lambda cold starts.

AJ Stuyvenberg, Datadog

Sessions related to event-driven architecture

API302 | Building next gen applications with event driven architecture
Learn about common integration patterns and discover how you can use AWS messaging services to connect microservices and coordinate data flow using minimal custom code. Learn and plan for idempotency, handling duplicating events and building resiliency into your architectures.

Eric Johnson, Principal Developer Advocate, AWS

API303 | Navigating the journey of serverless event-driven architecture
Learn about the journey businesses undertake when adopting EDAs, from initial design and implementation to ongoing operation and maintenance. The session highlights the many benefits EDAs can offer organizations and focuses on areas of EDA that are challenging and often overlooked. Through a combination of patterns, best practices, and practical tips, this session provides a comprehensive overview of the opportunities and challenges of implementing EDAs and helps you understand how you can use them to drive business success.

David Boyne, Senior Developer Advocate, AWS

API309 | Advanced integration patterns and trade-offs for loosely coupled apps
In this session, learn about common design trade-offs for distributed systems, how to navigate them with design patterns, and how to embed those patterns in your cloud automation.

Dirk Fröhner, Principal Solutions Architect, AWS
Gregor Hohpe, Senior Principal Evangelist, AWS

SVS205 | Getting started building serverless event-driven applications
Learn about the process of prototyping a solution from concept to a fully featured application that uses Amazon API Gateway, AWS Lambda, Amazon EventBridge, AWS Step Functions, Amazon DynamoDB, AWS Application Composer, and more. Learn why serverless is a great tool set for experimenting with new ideas and how the extensibility and modularity of serverless applications allow you to start small and quickly make your idea a reality.

Emily Shea, Head of Application Integration Go-to-Market, AWS
Naren Gakka, Solutions Architect, AWS

API206 | Bringing workloads together with event-driven architecture
Attend this session to learn the steps to bring your existing container workloads closer together using event-driven architecture with minimal code changes and a high degree of reusability. Using a real-life business example, this session walks through a demo to highlight the power of this approach.

Dhiraj Mahapatro, Principal Solutions Architect, AWS
Nicholas Stumpos, JPMorgan Chase & Co

COM301 | Advanced event-driven patterns with Amazon EventBridge
Gain an understanding of the characteristics of EventBridge and how it plays a pivotal role in serverless architectures. Learn the primary elements of event-driven architecture and some of the best practices. With real-world use cases, explore how the features of EventBridge support implementing advanced architectural patterns in serverless.

Sheen Brisals, The LEGO Group

Sessions related to serverless APIs

SVS301 | Building APIs: Choosing the best API solution and strategy for your workloads
Learn about access patterns and how to evaluate the best API technology for your applications. The session considers the features and benefits of Amazon API Gateway, AWS AppSync, Amazon VPC Lattice, and other options.

Josh Kahn, Tech Leader Serverless, AWS
Arthi Jaganathan, Principal Solutions Architect, AWS

SVS323 | I didn’t know Amazon API Gateway did that
This session provides an introduction to Amazon API Gateway and the problems it solves. Learn about the moving parts of API Gateway and how it works, including common and not-so-common use cases. Discover why you should use API Gateway and what it can do.

Eric Johnson, Principal Developer Advocate, AWS

FWM201 | What’s new with AWS AppSync for enterprise API developers
Join this session to learn about all the exciting new AWS AppSync features released this year that make it even more seamless for API developers to realize the benefits of GraphQL for application development.

Michael Liendo, Senior Developer Advocate, AWS
Brice Pellé, Principal Product Manager, AWS

FWM204 | Implement real-time event patterns with WebSockets and AWS AppSync
Learn how the PGA Tour uses AWS AppSync to deliver real-time event updates to their app users; review new features, like enhanced filtering options and native integration with Amazon EventBridge; and provide a sneak peek at what’s coming next.

Ryan Yanchuleff, Senior Solutions Architect, AWS
Bill Fine, Senior Product Manager, AWS
David Provan, PGA Tour

Sessions related to AWS Step Functions

API401 | Advanced workflow patterns and business processes with AWS Step Functions
Learn about architectural best practices and repeatable patterns for building workflows and cost optimizations, and discover handy cheat codes that you can use to build secure, high-scale, high-performance serverless applications

Ben Smith, Principal Developer Advocate, AWS

BOA304 | Using AI and serverless to automate video production
Learn how to use Step Functions to build workflows using AI services and how to use Amazon EventBridge real-time events.

Marcia Villalba, Principal Developer Advocate, AWS

SVS204 | Building Serverlesspresso: Creating event-driven architectures
This session explores the design decisions that were made when building Serverlesspresso, how new features influenced the development process, and lessons learned when creating a production-ready application using this approach. Explore useful patterns and options for extensibility that helped in the design of a robust, scalable solution that costs about one dollar per day to operate. This session includes examples you can apply to your serverless applications and complex architectural challenges for larger applications.

James Beswick, Senior Manager Developer Advocacy, AWS

API310 | Scale interactive data analysis with Step Functions Distributed Map
Learn how to build a data processing or other automation once and readily scale it to thousands of parallel processes with serverless technologies. Explore how this approach simplifies development and error handling while improving speed and lowering cost. Hear from an AWS customer that refactored an existing machine learning application to use Distributed Map and the lessons they learned along the way.

Adam Wagner, Principal Solutions Architect, AWS
Roberto Iturralde, Vertex Pharmaceuticals

Sessions related to handling data using serverless services and serverless databases

SVS307 | Scaling your serverless data processing with Amazon Kinesis and Kafka
Explore how to build scalable data processing applications using AWS Lambda. Learn practical insights into integrating Lambda with Amazon Kinesis and Apache Kafka using their event-driven models for real-time data streaming and processing.

Julian Wood, Principal Developer Advocate, AWS

DAT410 | Advanced data modeling with Amazon DynamoDB
This session shows you advanced techniques to get the most out of DynamoDB. Learn how to “think in DynamoDB” by learning the DynamoDB foundations and principles for data modeling. Learn practical strategies and DynamoDB features to handle difficult use cases in your application.

Alex De Brie – Independent consultant

COM308 | Serverless data streaming: Amazon Kinesis Data Streams and AWS Lambda
Explore the intricacies of creating scalable, production-ready data streaming architectures using Kinesis Data Streams and Lambda. Delve into tips and best practices essential to navigating the challenges and pitfalls inherent to distributed systems that arise along the way, and observe how AWS services work and interact.

Anahit Pogosova, Solita

Additional resources

If you are attending the event, there are many chalk talks, workshops, and other sessions to visit. See ServerlessLand for a full list of all the serverless sessions and also the Serverless Hero, Danielle Heberling’s Serverless re:Invent attendee guide for her top picks.

Visit us in the AWS Village in the Expo Hall where you can find the Serverless and Containers booth and enjoy a free cup of coffee at Serverlesspresso.

For more serverless learning resources, visit Serverless Land.

Enhanced Amazon CloudWatch metrics for Amazon EventBridge

2023-11-11 James Beswick

Post Syndicated from James Beswick original https://aws.amazon.com/blogs/compute/enhanced-amazon-cloudwatch-metrics-for-amazon-eventbridge/

This post is written by Vaibhav Shah, Sr. Solutions Architect.

Customers use event-driven architectures to orchestrate and automate their event flows from producers to consumers. Amazon EventBridge acts as a serverless event router for various targets based on event rules. It decouples the producers and consumers, allowing customers to build asynchronous architectures.

EventBridge provides metrics to enable you to monitor your events. Some of the metrics include: monitoring the number of partner events ingested, the number of invocations that failed permanently, and the number of times a target is invoked by a rule in response to an event, or the number of events that matched with any rule.

In response to customer requests, EventBridge has added additional metrics that allow customers to monitor their events and provide additional visibility. This blog post explains these new capabilities.

What’s new?

EventBridge has new metrics mainly around the API, events, and invocations metrics. These metrics give you insights into the total number of events published, successful events published, failed events, number of events matched with any or specific rule, events rejected because of throttling, latency, and invocations based metrics.

This allows you to track the entire span of event flow within EventBridge and quickly identify and resolve issues as they arise.

EventBridge now has the following metrics:

Metric	Description	Dimensions and Units
PutEventsLatency	The time taken per PutEvents API operation	None Units: Milliseconds
PutEventsRequestSize	The size of the PutEvents API request in bytes	None Units: Bytes
MatchedEvents	Number of events that matched with any rule, or a specific rule	None RuleName, EventBusName, EventSourceName Units: Count
ThrottledRules	The number of times rule execution was throttled.	None, RuleName Unit: Count
PutEventsApproximateCallCount	Approximate total number of calls in PutEvents API calls.	None Units: Count
PutEventsApproximateThrottledCount	Approximate number of throttled requests in PutEvents API calls.	None Units: Count
PutEventsApproximateFailedCount	Approximate number of failed PutEvents API calls.	None Units: Count
PutEventsApproximateSuccessCount	Approximate number of successful PutEvents API calls.	None Units: Count
PutEventsEntriesCount	The number of event entries contained in a PutEvents request.	None Units: Count
PutEventsFailedEntriesCount	The number of event entries contained in a PutEvents request that failed to be ingested.	None Units: Count
PutPartnerEventsApproximateCallCount	Approximate total number of calls in PutPartnerEvents API calls. (visible in Partner’s account)	None Units: Count
PutPartnerEventsApproximateThrottledCount	Approximate number of throttled requests in PutPartnerEvents API calls. (visible in Partner’s account)	None Units: Count
PutPartnerEventsApproximateFailedCount	Approximate number of failed PutPartnerEvents API calls. (visible in Partner’s account)	None Units: Count
PutPartnerEventsApproximateSuccessCount	Approximate number of successful PutPartnerEvents API calls. (visible in Partner’s account)	None Units: Count
PutPartnerEventsEntriesCount	The number of event entries contained in a PutPartnerEvents request.	None Units: Count
PutPartnerEventsFailedEntriesCount	The number of event entries contained in a PutPartnerEvents request that failed to be ingested.	None Units: Count
PutPartnerEventsLatency	The time taken per PutPartnerEvents API operation (visible in Partner’s account)	None Units: Milliseconds
InvocationsCreated	Number of times a target is invoked by a rule in response to an event. One invocation attempt represents a single count for this metric.	None Units: Count
InvocationAttempts	Number of times EventBridge attempted invoking a target.	None Units: Count
SuccessfulInvocationAttempts	Number of times target was successfully invoked.	None Units: Count
RetryInvocationAttempts	The number of times a target invocation has been retried.	None Units: Count
IngestiontoInvocationStartLatency	The time to process events, measured from when an event is ingested by EventBridge to the first invocation of a target.	None, RuleName, EventBusName Units: Milliseconds
IngestiontoInvocationCompleteLatency	The time taken from event Ingestion to completion of the first successful invocation attempt	None, RuleName, EventBusName Units: Milliseconds

Use-cases for these metrics

These new metrics help you improve observability and monitoring of your event-driven applications. You can proactively monitor metrics that help you understand the event flow, invocations, latency, and service utilization. You can also set up alerts on specific metrics and take necessary actions, which help improve your application performance, proactively manage quotas, and improve resiliency.

Monitor service usage based on Service Quotas

The PutEventsApproximateCallCount metric in the events family helps you identify the approximate number of events published on the event bus using the PutEvents API action. The PutEventsApproximateSuccessfulCount metric shows the approximate number of successful events published on the event bus.

Similarly, you can monitor throttled and failed events count with PutEventsApproximateThrottledCount and PutEventsApproximateFailedCount respectively. These metrics allow you to monitor if you are reaching your quota for PutEvents. You can use a CloudWatch alarm and set a threshold close to your account quotas. If that is triggered, send notifications using Amazon SNS to your operations team. They can work to increase the Service Quotas.

You can also set an alarm on the PutEvents throttle limit in transactions per second service quota.

Navigate to the Service Quotas console. On the left pane, choose AWS services, search for EventBridge, and select Amazon EventBridge (CloudWatch Events).
In the Monitoring section, you can monitor the percentage utilization of the PutEvents throttle limit in transactions per second.
Go to the Alarms tab, and choose Create alarm. In Alarm threshold, choose 80% of the applied quota value from the dropdown. Set the Alarm name to PutEventsThrottleAlarm, and choose Create.
To be notified if this threshold is breached, navigate to Amazon CloudWatch Alarms console and choose PutEventsThrottleAlarm.
Select the Actions dropdown from the top right corner, and choose Edit.
On the Specify metric and conditions page, under Conditions, make sure that the Threshold type is selected as Static and the % Utilization selected as Greater/Equal than 80. Choose Next.
Configure actions to send notifications to an Amazon SNS topic and choose Next.
The Alarm name should be already set to PutEventsThrottleAlarm. Choose Next, then choose Update alarm.

This helps you get notified when the percentage utilization of PutEvents throttle limit in transactions per second reaches close to the threshold set. You can then request Service Quota increases if required.

Similarly, you can also create CloudWatch alarms on percentage utilization of Invocations throttle limit in transactions per second against the service quota.

Enhanced observability

The PutEventsLatency metric shows the time taken per PutEvents API operation. There are two additional metrics, IngestiontoInvocationStartLatency metric and IngestiontoInvocationCompleteLatency metric. The first metric shows the time to process events measured from when the events are first ingested by EventBridge to the first invocation of a target. The second shows the time taken from event ingestion to completion of the first successful invocation attempt.

This helps identify latency-related issues from the time of ingestion until the time it reaches the target based on the RuleName. If there is high latency, these two metrics give you visibility into this issue, allowing you to take appropriate action.

You can set a threshold around these metrics, and if the threshold is triggered, the defined actions can help recover from potential failures. One of the defined actions here can be to send events generated later to EventBridge in the secondary Region using EventBridge global endpoints.

Sometimes, events are not delivered to the target specified in the rule. This can be because the target resource is unavailable, you don’t have permission to invoke the target, or there are network issues. In such scenarios, EventBridge retries to send these events to the target for 24 hours or up to 185 times, both of which are configurable.

The new RetryInvocationAttempts metric shows the number of times the EventBridge has retried to invoke the target. The retries are done when requests are throttled, target service having availability issues, network issues, and service failures. This provides additional observability to the customers and can be used to trigger a CloudWatch alarm to notify teams if the desired threshold is crossed. If the retries are exhausted, store the failed events in the Amazon SQS dead-letter queues to process failed events for the later time.

In addition to these, EventBridge supports additional dimensions like DetailType, Source, and RuleName to MatchedEvents metrics. This helps you monitor the number of matched events coming from different sources.

Navigate to the Amazon CloudWatch. On the left pane, choose Metrics, and All metrics.
In the Browse section, select Events, and Source.
From the Graphed metrics tab, you can monitor matched events coming from different sources.

Failover events to secondary Region

The PutEventsFailedEntriesCount metric shows the number of events that failed ingestion. Monitor this metric and set a CloudWatch alarm. If it crosses a defined threshold, you can then take appropriate action.

Also, set an alarm on the PutEventsApproximateThrottledCount metric, which shows the number of events that are rejected because of throttling constraints. For these event ingestion failures, the client must resend the failed events to the event bus again, allowing you to process every single event critical for your application.

Alternatively, send events to EventBridge service in the secondary Region using Amazon EventBridge global endpoints to improve resiliency of your event-driven applications.

Conclusion

This blog shows how to use these new metrics to improve the visibility of event flows in your event-driven applications. It helps you monitor the events more effectively, from invocation until the delivery to the target. This improves observability by proactively alerting on key metrics.

For more serverless learning resources, visit Serverless Land.

Introducing the Amazon Linux 2023 runtime for AWS Lambda

2023-11-10 James Beswick

Post Syndicated from James Beswick original https://aws.amazon.com/blogs/compute/introducing-the-amazon-linux-2023-runtime-for-aws-lambda/

This post is written by Rakshith Rao, Senior Solutions Architect.

AWS Lambda now supports Amazon Linux 2023 (AL2023) as a managed runtime and container base image. Named provided.al2023, this runtime provides an OS-only environment to run your Lambda functions.

It is based on the Amazon Linux 2023 minimal container image release and has several improvements over Amazon Linux 2 (AL2), such as a smaller deployment footprint, updated versions of libraries like glibc, and a new package manager.

What are OS-only Lambda runtimes?

Lambda runtimes define the execution environment where your function runs. They provide the OS, language support, and additional settings such as environment variables and certificates.

Lambda provides managed runtimes for Java, Python, Node.js, .NET, and Ruby. However, if you want to develop your Lambda functions in programming languages that are not supported by Lambda’s managed language runtimes, the ‘provided’ runtime family provides an OS-only environment in which you can run code written in any language. This release extends the provided runtime family to support Amazon Linux 2023.

Customers use these OS-only runtimes in three common scenarios. First, they are used with languages that compile to native code, such as Go, Rust, C++, .NET Native AOT and Java GraalVM Native. Since you only upload the compiled binary to Lambda, these languages do not require a dedicated language runtime, they only require an OS environment in which the binary can run.

Second, the OS-only runtimes also enable building third-party language runtimes that you can use off the shelf. For example, you can write Lambda functions in PHP using Bref, or Swift using the Swift AWS Lambda Runtime.

Third, you can use the OS-only runtime to deploy custom runtimes, which you build for a language or language version which Lambda does not provide a managed runtime. For example, Node.js 19 – Lambda only provides managed runtimes for LTS releases, which for Node.js are the even-numbered releases.

New in Amazon Linux 2023 base image for Lambda

Updated packages

AL2023 base image for Lambda is based on the AL2023-minimal container image and includes various package updates and changes compared with provided.al2.

The version of glibc in the AL2023 base image has been upgraded to 2.34, from 2.26 that was bundled in the AL2 base image. Some libraries that developers wanted to use in provided runtimes required newer versions of glibc. With this launch, you can now use an up-to-date version of glibc with your Lambda function.

The AL2 base image for Lambda came pre-installed with Python 2.7. This was needed because Python was a required dependency for some of the packages that were bundled in the base image. The AL2023 base image for Lambda has removed this dependency on Python 2.7 and does not come with any pre-installed language runtime. You are free to choose and install any compatible Python version that you need.

Since the AL2023 base image for Lambda is based on the AL2023-minimal distribution, you also benefit from a significantly smaller deployment footprint. The new image is less than 40MB compared to the AL2-based base image, which is over 100MB in size. You can find the full list of packages available in the AL2023 base image for Lambda in the “minimal container” column of the AL2023 package list documentation.

Package manager

Amazon Linux 2023 uses dnf as the package manager, replacing yum, which was the default package manager in Amazon Linux 2. AL2023 base image for Lambda uses microdnf as the package manager, which is a standalone implementation of dnf based on libdnf and does not require extra dependencies such as Python. microdnf in provided.al2023 is symlinked as dnf. Note that microdnf does not support all options of dnf. For example, you cannot install a remote rpm using the rpm’s URL or install a local rpm file. Instead, you can use the rpm command directly to install such packages.

This example Dockerfile shows how you can install packages using dnf while building a container-based Lambda function:

# Use the Amazon Linux 2023 Lambda base image
FROM public.ecr.aws/lambda/provided.al2023

# Install the required Python version
RUN dnf install -y python3

Runtime support

With the launch of provided.al2023 you can migrate your AL2 custom runtime-based Lambda functions right away. It also sets the foundation of future Lambda managed runtimes. The future releases of managed language runtimes such as Node.js 20, Python 3.12, Java 21, and .NET 8 are based on Amazon Linux 2023 and will use provided.al2023 as the base image.

Changing runtimes and using other compute services

Previously, the provided.al2 base image was built as a custom image that used a selection of packages from AL2. It included packages like curl and yum that were needed to build functions using custom runtime. Also, each managed language runtime used different packages based on the use case.

Since future releases of managed runtimes use provided.al2023 as the base image, they contain the same set of base packages that come with AL2023-minimal. This simplifies migrating your Lambda function from a custom runtime to a managed language runtime. It also makes it easier to switch to other compute services like AWS Fargate or Amazon Elastic Container Services (ECS) to run your application.

Upgrading from AL1-based runtimes

For more information on Lambda runtime deprecation, see Lambda runtimes.

AL1 end of support is scheduled for December 31, 2023. The AL1-based runtimes go1.x, java8 and provided will be deprecated from this date. You should migrate your Go based Lambda functions to the provided runtime family, such as provided.al2 or provided.al2023. Using a provided runtime offers several benefits over the go1.x runtime. First, you can run your Lambda functions on AWS Graviton2 processors that offer up to 34% better price-performance compared to functions running on x86_64 processors. Second, it offers a smaller deployment package and faster function invoke path. And third, it aligns Go with other languages that also compile to native code and run on the provided runtime family.

The deprecation of the Amazon Linux 1 (AL1) base image for Lambda is also scheduled for December 31, 2023. With provided.al2023 now generally available, you should start planning the migration of your go1.x and AL1 based Lambda functions to provided.al2023.

Using the AL2023 base image for Lambda

To build Lambda functions using a custom runtime, follow these steps using the provided.al2023 runtime.

AWS Management Console

Navigate to the Create function page in the Lambda console. To use the AL2023 custom runtime, select Provide your own bootstrap on Amazon Linux 2023 as the Runtime value:

AWS Serverless Application Model (AWS SAM) template

If you use the AWS SAM template to build and deploy your Lambda function, use the provided.al2023 as the value of the Runtime:

Resources:
  HelloWorldFunction:
    Type: AWS::Serverless::Function
    Properties:
      CodeUri: hello-world/
      Handler: my.bootstrap.file
      Runtime: provided.al2023

Building Lambda functions that compile natively

Lambda’s custom runtime simplifies the experience to build functions in languages that compile to native code, broadening the range of languages you can use. Lambda provides the Runtime API, an HTTP API that custom runtimes can use to interact with the Lambda service. Implementations of this API, called Runtime Interface Client (RIC), allow your function to receive invocation events from Lambda, send the response back to Lambda, and report errors to the Lambda service. RICs are available as language-specific libraries for several popular programming langauges such as Go, Rust, Python, and Java.

For example, you can build functions using Go as shown in the Building with Go section of the Lambda developer documentation. Note that the name of the executable file of your function should always be bootstrap in provided.al2023 when using the zip deployment model. To use AL2023 in this example, use provided.al2023 as the runtime for your Lambda function.

If you are using CLI set the --runtime option to provided.al2023:

aws lambda create-function --function-name myFunction \
--runtime provided.al2023 --handler bootstrap \
--role arn:aws:iam::111122223333:role/service-role/my-lambda-role \
--zip-file fileb://myFunction.zip

If you are using AWS Serverless Application Model, use provided.al2023 as the value of the Runtime in your AWS SAM template file:

AWSTemplateFormatVersion: '2010-09-09'
Transform: 'AWS::Serverless-2016-10-31'
Resources:
  HelloWorldFunction:
    Type: AWS::Serverless::Function
    Metadata:
      BuildMethod: go1.x
    Properties:
      CodeUri: hello-world/ # folder where your main program resides
      Handler: bootstrap
      Runtime: provided.al2023
      Architectures: [arm64]

If you run your function as a container image as shown in the Deploy container image example, use this Dockerfile. You can use any name for the executable file of your function when using container images. You need to specify the name of the executable as the ENTRYPOINT in your Dockerfile:

FROM golang:1.20 as build
WORKDIR /helloworld

# Copy dependencies list
COPY go.mod go.sum ./

# Build with optional lambda.norpc tag
COPY main.go .
RUN go build -tags lambda.norpc -o main main.go

# Copy artifacts to a clean image
FROM public.ecr.aws/lambda/provided:al2023
COPY --from=build /helloworld/main ./main
ENTRYPOINT [ "./main" ]

Conclusion

With this launch, you can now build your Lambda functions using Amazon Linux 2023 as the custom runtime or use it as the base image to run your container-based Lambda functions. You benefit from the updated versions of libraries such as glibc, new package manager, and smaller deployment size than Amazon Linux 2 based runtimes. Lambda also uses Amazon Linux 2023-minimal as the basis for future Lambda runtime releases.

For more serverless learning resources, visit Serverless Land.

Scaling improvements when processing Apache Kafka with AWS Lambda

2023-11-10 Julian Wood

Post Syndicated from Julian Wood original https://aws.amazon.com/blogs/compute/scaling-improvements-when-processing-apache-kafka-with-aws-lambda/

AWS Lambda is improving the automatic scaling behavior when processing data from Apache Kafka event-sources. Lambda is increasing the default number of initial consumers, improving how quickly consumers scale up, and helping to ensure that consumers don’t scale down too quickly. There is no additional action that you must take, and there is no additional cost.

Running Kafka on AWS

Apache Kafka is a popular open-source platform for building real-time streaming data pipelines and applications. You can deploy and manage your own Kafka solution on-premises or in the cloud on Amazon EC2.

Amazon Managed Streaming for Apache Kafka (MSK) is a fully managed service that makes it easier to build and run applications that use Kafka to process streaming data. MSK Serverless is a cluster type for Amazon MSK that allows you to run Kafka without having to manage and scale cluster capacity. It automatically provisions and scales capacity while managing the partitions in your topic, so you can stream data without thinking about right-sizing or scaling clusters. MSK Serverless offers a throughput-based pricing model, so you pay only for what you use. For more information, see Using Kafka to build your streaming application.

Using Lambda to consume records from Kafka

Processing streaming data can be complex in traditional, server-based architectures, especially if you must react in real-time. Many organizations spend significant time and cost managing and scaling their streaming platforms. In order to react fast, they must provision for peak capacity, which adds complexity.

Lambda and serverless architectures remove the undifferentiated heavy lifting when processing Kafka streams. You don’t have to manage infrastructure, can reduce operational overhead, lower costs, and scale on-demand. This helps you focus more on building streaming applications. You can write Lambda functions in a number of programming languages, which provide flexibility when processing streaming data.

Lambda event source mapping

Lambda can integrate natively with your Kafka environments as a consumer to process stream data as soon as it’s generated.

To consume streaming data from Kafka, you configure an event source mapping (ESM) on your Lambda functions. This is a resource managed by the Lambda service, which is separate from your function. It continually polls records from the topics in the Kafka cluster. The ESM optionally filters records and batches them into a payload. Then, it calls the Lambda Invoke API to deliver the payload to your Lambda function synchronously for processing.

As Lambda manages the pollers, you don’t need to manage a fleet of consumers across multiple teams. Each team can create and configure their own ESM with Lambda handling the polling.

AWS Lambda event source mapping

For more information on using Lambda to process Kafka streams, see the learning guide.

Scaling and throughput

Kafka uses partitions to increase throughput and spread the load of records to all brokers in a cluster.

The Lambda event source mapping resource includes pollers and processors. Pollers have consumers that read records from Kafka partitions. The poller assigners send them to processors which batch the records and invoke your function.

When you create a Kafka event source mapping, Lambda allocates consumers to process all partitions in the Kafka topic. Previously, Lambda allocated a minimum of one processor for a consumer.

Lambda previous initial scaling

With these scaling improvements, Lambda allocates multiple processors to improve processing. This reduces the possibility of a single invoke slowing down the entire processing stream.

Lambda updated initial scaling

Each consumer sends records to multiple processors running in parallel to handle increased workloads. Records in each partition are only assigned to a single processor to maintain order.

Lambda automatically scales up or down the number of consumers and processors based on workload. Lambda samples the consumer offset lag of all the partitions in the topic every minute. If the lag is increasing, this means Lambda can’t keep up with processing the records from the partition.

The scaling algorithm accounts for the current offset lag, and also the rate of new messages added to the topic. Lambda can reach the maximum number of consumers within three minutes to lower the offset lag as quickly as possible. Lambda is also reducing the scale down behavior to ensure records are processed more quickly and latency is reduced, particularly for bursty workloads.

Total processors for all pollers can only scale up to the total number of partitions in the topic.

After successful invokes, the poller periodically commits offsets to the respective brokers.

Lambda further scaling

You can monitor the throughput of your Kafka topic using consumer metrics consumer_lag and consumer_offset.

To check how many function invocations occur in parallel, you can also monitor the concurrency metrics for your function. The concurrency is approximately equal to the total number of processors across all pollers, depending on processor activity. For example, if three pollers have five processors running for a given ESM, the function concurrency would be approximately 15 (5 + 5 + 5).

Seeing the improved scaling in action

There are a number of Serverless Patterns that you can use to process Kafka streams using Lambda. To set up Amazon MSK Serverless, follow the instructions in the GitHub repo:

Create an example Amazon MSK Serverless topic with 1000 partitions.

./kafka-topics.sh --create --bootstrap-server "{bootstrap-server}" --command-config client.properties --replication-factor 3 --partitions 1000 --topic msk-1000p

Add records to the topic using a UUID as a key to distribute records evenly across partitions. This example adds 13 million records.

for x in {1..13000000}; do echo $(uuidgen -r),message_$x; done | ./kafka-console-producer.sh --broker-list "{bootstrap-server}" --topic msk-1000p --producer.config client.properties --property parse.key=true --property key.separator=, --producer-property acks=all

Create a Python function based on this pattern, which logs the processed records.
Amend the function code to insert a delay of 0.1 seconds to simulate record processing.

import json
import base64
import time

def lambda_handler(event, context):
    # Define a variable to keep track of the number of the message in the batch of messages
    i=1
    # Looping through the map for each key (combination of topic and partition)
    for record in event['records']:
        for messages in event['records'][record]:
            print("********************")
            print("Record number: " + str(i))
            print("Topic: " + str(messages['topic']))
            print("Partition: " + str(messages['partition']))
            print("Offset: " + str(messages['offset']))
            print("Timestamp: " + str(messages['timestamp']))
            print("TimestampType: " + str(messages['timestampType']))
            if None is not messages.get('key'):
                b64decodedKey=base64.b64decode(messages['key'])
                decodedKey=b64decodedKey.decode('ascii')
            else:
                decodedKey="null"
            if None is not messages.get('value'):
                b64decodedValue=base64.b64decode(messages['value'])
                decodedValue=b64decodedValue.decode('ascii')
            else:
                decodedValue="null"
            print("Key = " + str(decodedKey))
            print("Value = " + str(decodedValue))
            i=i+1
            time.sleep(0.1)
    return {
        'statusCode': 200,
    }

Configure the ESM to point to the previously created cluster and topic.
Use the default batch size of 100. Set the StartingPosition to TRIM_HORIZON to process from the beginning of the stream.
Deploy the function, which also adds and configures the ESM.
View the Amazon CloudWatch ConcurrentExecutions and OffsetLag metrics to view the processing.

With the scaling improvements, once the ESM is configured, the ESM and function scale up to handle the number of partitions.

Lambda automatic scaling improvement graph

Increasing data processing throughput

It is important that your function can keep pace with the rate of traffic. A growing offset lag means that the function processing cannot keep up. If the age is high in relation to the stream’s retention period, you can lose data as records expire from the stream.

This value should generally not exceed 50% of the stream’s retention period. When the value reaches 100% of the stream retention period, data is lost. One temporary solution is to increase the retention time of the stream. This gives you more time to resolve the issue before losing data.

There are several ways to improve processing throughput.

Avoid processing unnecessary records by using content filtering to control which records Lambda sends to your function. This helps reduce traffic to your function, simplifies code, and reduces overall cost.
Lambda allocates processors across all pollers based on the number of partitions up to a maximum of one concurrent Lambda function per partition. You can increase the number of processing Lambda functions by increasing the number of partitions.
For compute intensive functions, you can increase the memory allocated to your function, which also increases the amount of virtual CPU available. This can help reduce the duration of a processing function.
Lambda polls Kafka with a configurable batch size of records. You can increase the batch size to process more records in a single invocation. This can improve processing time and reduce costs, particularly if your function has an increased init time. A larger batch size increases the latency to process the first record in the batch, but potentially decreases the latency to process the last record in the batch. There is a tradeoff between cost and latency when optimizing a partition’s capacity and the decision depends on the needs of your workload.
Ensure that your producers evenly distribute records across partitions using an effective partition key strategy. A workload is unbalanced when a single key dominates other keys, creating a hot partition, which impacts throughput.

See Increasing data processing throughput for some additional guidance.

Conclusion

Today, AWS Lambda is improving the automatic scaling behavior when processing data from Apache Kafka event-sources. Lambda is increasing the default number of initial consumers, improving how quickly they scale up, and ensuring they don’t scale down too quickly. There is no additional action that you must take, and there is no additional cost.

You can explore the scaling improvements with your existing workloads or deploy an Amazon MSK cluster and try one of the patterns to measure processing time.

To explore using Lambda to process Kafka streams, see the learning guide.

For more serverless learning resources, visit Serverless Land.

How Gilead used Amazon Redshift to quickly and cost-effectively load third-party medical claims data

2023-11-08 Rajiv Arora

Post Syndicated from Rajiv Arora original https://aws.amazon.com/blogs/big-data/how-gilead-used-amazon-redshift-to-quickly-and-cost-effectively-load-third-party-medical-claims-data/

This post was co-written with Rajiv Arora, Director of Data Science Platform at Gilead Life Sciences.

Gilead Sciences, Inc. is a biopharmaceutical company committed to advancing innovative medicines to prevent and treat life-threatening diseases, including HIV, viral hepatitis, inflammation, and cancer. A leader in virology, Gilead historically relied on these drugs for growth but now through strategic investments, Gilead is expanding and increasing their focus in oncology, having acquired Kite and Immunomedics to boost their exposure to cell therapy and non-cell therapy, making it the primary growth engine. Because Gilead is expanding into biologics and large molecule therapies, and has an ambitious goal of launching 10 innovative therapies by 2030, there is heavy emphasis on using data with AI and machine learning (ML) to accelerate the drug discovery pipeline.

Amazon Redshift Serverless is a fully managed cloud data warehouse that allows you to seamlessly create your data warehouse with no infrastructure management required. You pay only for the compute resources and storage that you use. Redshift Serverless measures data warehouse capacity in Redshift Processing Units (RPUs), which are part of the compute resources. All of the data stored in your warehouse, such as tables, views, and users, make up a namespace in Redshift Serverless.

One of the benefits of Redshift Serverless is that you don’t need to size your data warehouse for your peak workload. The peak workload includes loading periodic large datasets in multi-terabyte range. You can set a base RPU from 8 up to 512 and Redshift Serverless will automatically scale the RPUs to meet your workload demands. This makes it straightforward to manage your data warehouse in a cost-effective manner.

In this post, we share how Gilead collaborated with AWS to redesign their data ingestion process. They used Redshift Serverless as their data producer to load third-party medical claims data in a fast and cost-effective way, reducing load times from days to hours.

Gilead use case

Gilead loads a variety of data from hundreds of sources to their R&D data environment. They recently needed to do a monthly load of 140 TB of uncompressed healthcare claims data in under 24 hours after receiving it to provide analysts and data scientists with up-to-date information on a patient’s healthcare journey. This data volume is expected to increase monthly and is fully refreshed each month. The 3-node RA3 16XL provisioned cluster that had previously been hosting their warehouse was taking around 12 hours to ingest this data to Amazon Redshift, and Gilead was looking to optimize the data ingestion process in a more dynamic manner. Working with Amazon Redshift specialists from AWS, Gilead chose Redshift Serverless as a way to cost-effectively load this data and then use Redshift data sharing to share the final dataset to two additional Redshift data warehouses for end-user queries.

Loading data is a key process for any analytical system, including Amazon Redshift. When loading very large datasets, it’s important to not only load the data as quickly as possible but also in a way that optimizes the consumption queries.

Gilead’s healthcare claims data took 40 hours to load, which meant delays in using the data for downstream processes. The teams sought improvements, targeting a maximum 24-hour SLA for the load. They achieved the load in 8 hours, an 80% reduction in time to make data available.

Solution overview

After collaborating, the Gilead and AWS teams decided on a two-step process to load the data to Amazon Redshift. First, the data was loaded without a distkey and sortkey, which let the load process use the full parallel resources of the cluster. Then we used a deep copy to redistribute this data and add the desired distribution and sort characteristics.

The solution uses Redshift Serverless. The team wanted to ingest data to meet the required SLA, and the following approaches were benchmarked:

COPY command – The COPY command uses the Amazon Redshift massively parallel processing (MPP) architecture to read and load data in parallel from files on Amazon Simple Storage Service (Amazon S3)
Data lake analytics – Amazon Redshift Spectrum is used to query data directly from files on Amazon S3 by selecting a subset of columns and avoiding the intermediate step of copying data to staging table

Initial Solution approach: Single COPY command

The team determined it would be more effective to apply the distribution and sort keys in a post-copy step. The data was loaded first using automatic distribution of data. This took roughly 12 hours to complete. The team created open and closed claims tables with defined dist keys and with 20% of the columns to alleviate the need to query the larger table. With this success, we learned that we can still improve the big copy, as detailed in the following sections.

Proposed Solution approach 1: Parallel COPY command

Based on the initial solution approach above, the team tested yearly parallel copy commands as illustrated in the following diagram.

Below are the findings and learnings from this approach:

Ingesting data for 4 years using parallel copy showed a 25% performance improvement over the single copy command.
Compared to Initial solution approach, where we were taking 12 hours to ingest the data, we further optimized this runtime by 67% by segregating the data ingestion into separate yearly staging tables and running parallel copy commands.
After the data was loaded into staging yearly tables, we created the open and closed claim tables with an auto distkey with the subset of columns required for larger reporting groups. It took an additional 1 hour to create.

The team used a manifest file to make sure that the COPY command loads all of the required files for the respective year for ingesting.

Proposed Solution approach 2: Data Lake analytics

The team used this approach with Redshift Spectrum to load only the required columns to Redshift Serverless, which avoided loading data into multiple yearly tables and directly to a single table. The following diagram illustrates this approach.

The workflow consists of the following steps:

Crawl the files using AWS Glue.
Create a data lake external schema and table in Redshift Serverless.
Create two separate claims table for open and closed claims because open claims are most frequently consumed and are 20% of the columns and 100% of the data.
Create open and closed tables with selective columns needed for optimal performance optimization during consumption instead of all columns in the original third-party dataset. The data volume distribution is as follows:
- Total number of open claims records = 50 billion
- Total number of closed claims records = 200 billion
- Overall, total number of records = 250 billion
Distribute open and closed tables with a customer-identified distkey.
Configure data ingestion into open and closed claims tables combined using Redshift Serverless with 512 RPUs. This took 1.5 hours, which is further improved by 70% compared to scenario 1. We chose 512 RPUs in order to load data in the fastest way possible.

In this method, data ingestion was streamlined by only loading essential fields from the medical claims dataset and by splitting the table into open and closed claims. Open claims data is most frequently accessed and constitutes only 20% of columns so by splitting the tables. The team not only improved the ingestion performance but also consumption.

Amazon Redshift recently launched automatic mounting of AWS Glue Data Catalog, making it easier to run data lake analytics without manually creating external schemas. You can query data lake tables directly from Amazon Redshift Query Editor v2 or your favorite SQL editors.

Recommendations and best practices

Consider the following recommendations when loading large-scale data in Amazon Redshift.

Use Redshift Serverless with maximum 512 RPUs to efficiently and quickly load data
Depending on consumption use case and query pattern, adopt either of the following approaches:
- When consumption queries require only selected fields from the dataset and most frequently access a subset of data, use data lake queries to load only the relevant columns from Amazon S3 into Amazon Redshift
- When consumption queries require all fields, use COPY commands with a manifest file to ingest data in parallel into multiple logically separated tables and create a database view with UNION ALL of all tables
Avoid using varchar(max) while creating tables and create VARCHAR columns with the right size

Final Architecture

The following diagram shows the high-level final architecture that was implemented.

Conclusion

With the scalability of Redshift Serverless, data sharing to decouple ingestion from consumption workloads, and data lake analytics to ingest data, Gilead made their 140 TB dataset available to their analysts within hours of it being delivered. The innovative architecture of using a serverless ingestion data warehouse, a serverless consumption data warehouse for power users, and their original 3-node provisioned cluster for standard queries gives Gilead isolation to ensure data loads don’t affect their users. The architecture provides scalability to serve infrequent large queries with their serverless consumer along with the benefit of a fixed-cost and fixed-performance option of their provisioned cluster for their standard user queries. Due to the monthly schedule of the data load and the variable need for large queries by consumers, Redshift Serverless proved to be a cost-effective option compared to simply increasing the provisioned cluster to serve each of these use cases.

This split producer/consumer model of using Redshift serverless can bring benefits to many workloads that have similar performance characteristics to Gilead’s warehouse. Customers regularly run large data loads infrequently, and those processes compete with user queries. With this pattern, you can rely on your queries to perform consistently regardless of whether new data is being loaded to the system. This strikes a balance between minimizing cost while maintaining performance and frees the system administrators to load data without affecting users.

About the Authors

Rajiv Arora is a Director of Clinical Data Science at Gilead Sciences with over 20 years of experience in the industry. He is responsible for the multi-modal data platform for the development organization and supports all statistical and predictive analytical infrastructure for RWE and Advanced Analytical functions.

Ritesh Kumar Sinha is an Analytics Specialist Solutions Architect based out of San Francisco. He has helped customers build scalable data warehousing and big data solutions for over 16 years. He loves to design and build efficient end-to-end solutions on AWS. In his spare time, he loves reading, walking, and doing yoga.

Raks Khare is an Analytics Specialist Solutions Architect at AWS based out of Pennsylvania. He helps customers architect data analytics solutions at scale on the AWS platform.

Brent Strong is a Senior Solutions Architect in the Healthcare and Life Sciences team at AWS. He has more than 15 years of experience in the industry, focusing on data and analytics and DevOps. At AWS, he works closely with large Life Sciences customers to help them deliver new and innovative treatments.

Phil Bates is a Senior Analytics Specialist Solutions Architect at AWS with over 25 years of data warehouse experience.

Introducing faster polling scale-up for AWS Lambda functions configured with Amazon SQS

2023-11-06 James Beswick

Post Syndicated from James Beswick original https://aws.amazon.com/blogs/compute/introducing-faster-polling-scale-up-for-aws-lambda-functions-configured-with-amazon-sqs/

This post was written by Anton Aleksandrov, Principal Solutions Architect, and Tarun Rai Madan, Senior Product Manager.

Today, AWS is announcing that AWS Lambda supports up to five times faster polling scale-up rate for spiky Lambda workloads configured with Amazon Simple Queue Service (Amazon SQS) as an event source.

This feature enables customers building event-driven applications using Lambda and SQS to achieve more responsive scaling during a sudden burst of messages in their SQS queues, and reduces the need to duplicate Lambda functions or SQS queues to achieve faster message processing.

Overview

Customers building modern event-driven and messaging applications with AWS Lambda use the Amazon SQS as a fundamental building block for creating decoupled architectures. Amazon SQS is a fully managed message queueing service for microservices, distributed systems, and serverless applications. When a Lambda function subscribes to an SQS queue as an event source, Lambda polls the queue, retrieves the messages, and sends retrieved messages in batches to the function handler for processing. To consume messages efficiently, Lambda detects the increase in queue depth, and increases the number of poller processes to process the queued messages.

Up until today, the Lambda was adding up to 60 concurrent executions per minute for Lambda functions subscribed to SQS queues, scaling up to a maximum of 1,250 concurrent executions in approximately 20 minutes. However, customers tell us that some of the modern event-driven applications they build using Lambda and SQS are sensitive to sudden spikes in messages, which may cause noticeable delay in processing of messages for end users. In order to harness the power of Lambda for applications that experience a burst of messages in SQS queues, these customers needed Lambda message polling to scale up faster.

With today’s announcement, Lambda functions that subscribe to an SQS queue can scale up to five times faster for queues that see a spike in message backlog, adding up to 300 concurrent executions per minute, and scaling up to a maximum of 1,250 concurrent executions. This scaling improvement helps to use the simplicity of Lambda and SQS integration to build event-driven applications that scale faster during a surge of incoming messages, particularly for real-time systems. It also offers customers the benefit of faster processing during spikes of messages in SQS queues, while continuing to offer the flexibility to limit the maximum concurrent Lambda invocations per SQS event source.

Controlling the maximum concurrent Lambda invocations by SQS

The new improved scaling rates are automatically applied to all AWS accounts using Lambda and SQS as an event source. There is no explicit action that you must take, and there’s no additional cost. This scaling improvement helps customers to build more performant Lambda applications where they need faster SQS polling scale-up. To prevent potentially overloading the downstream dependencies, Lambda provides customers the control to set the maximum number of concurrent executions at a function level with reserved concurrency, and event source level with maximum concurrency.

The following diagram illustrates settings that you can use to control the flow rate of an SQS event-source. You use reserved concurrency to control function-level scaling, and maximum concurrency to control event source scaling.

Reserved concurrency is the maximum concurrency that you want to allocate to a function. When a function has reserved concurrency allocated, no other functions can use that concurrency.

AWS recommends using reserved concurrency when you want to ensure a function has enough concurrency to scale up. When an SQS event source is attempting to scale up concurrent Lambda invocations, but the function has already reached the threshold defined by the reserved concurrency, the Lambda service throttles further function invocations.

This may result in SQS event source attempting to scale down, reducing the number of concurrently processed messages. Depending on the queue configuration, the throttled messages are returned to the queue for retrying, expire based on the retention policy, or sent to a dead-letter queue (DLQ) or on-failure destination.

The maximum concurrency setting allows you to control concurrency at the event source level. It allows you to define the maximum number of concurrent invocations the event source attempts to send to the Lambda function. For scenarios where a single function has multiple SQS event sources configured, you can define maximum concurrency for each event source separately, providing more granular control. When trying to add rate control to SQS event sources, AWS recommends you start evaluating maximum concurrency control first, as it provides greater flexibility.

Reserved concurrency and maximum concurrency are complementary capabilities, and can be used together. Maximum concurrency can help to prevent overwhelming downstream systems and throttled invocations. Reserved concurrency helps to ensure available concurrency for the function.

Example scenario

Consider your business must process large volumes of documents from storage. Once every few hours, your business partners upload large volumes of documents to S3 buckets in your account.

For resiliency, you’ve designed your application to send a message to an SQS queue for each of the uploaded documents, so you can efficiently process them without accidentally skipping any. The documents are processed using a Lambda function, which takes around two seconds to process a single document.

Processing these documents is a CPU-intensive operation, so you decide to process a single document per invocation. You want to use the power of Lambda to fan out the parallel processing to as many concurrent execution environments as possible. You want the Lambda function to scale up rapidly to process those documents in parallel as fast as possible, and scale-down to zero once all documents are processed to save costs.

When a business partner uploads 200,000 documents, 200,000 messages are sent to the SQS queue. The Lambda function is configured with an SQS event source, and it starts consuming the messages from the queue.

This diagram shows the results of running the test scenario before the SQS event source scaling improvements. As expected, you can see that concurrent executions grow by 60 per minute. It takes approximately 16 minutes to scale up to 900 concurrent executions gradually and process all the messages in the queue.

The following diagram shows the results of running the same test scenario after the SQS event source scaling improvements. The timeframe used for both charts is the same, but the performance on the second chart is better. Concurrent executions grow by 300 per minute. It only takes 4 minutes to scale up to 1,250 concurrent executions, and all the messages in the queue are processed in approximately 8 minutes.

Deploying this example

Use the example project to replicate this performance test in your own AWS account. Follow the instructions in README.md for provisioning the sample project in your AWS accounts using the AWS Cloud Development Kit (CDK).

This example project is configured to demonstrate a large-scale workload processing 200,000 messages. Running this sample project in your account may incur charges. See AWS Lambda pricing and Amazon SQS pricing.

Once deployed, use the application under the “sqs-cannon” directory to send 200,000 messages to the SQS queue (or reconfigure to any other number). It takes several minutes to populate the SQS queue with messages. After all messages are sent, enable the SQS event source, as described in the README.md, and monitor the charts in the provisioned CloudWatch dashboard.

The default concurrency quota for new AWS accounts is 1000. If you haven’t requested an increase in this quota, the number of concurrent executions is capped at this number. Use Service Quotas or contact your account team to request a concurrency increase.

Security best practices

Always use the least privileged permissions when granting your Lambda functions access to SQS queues. This reduces potential attack surface by ensuring that only specific functions have permissions to perform specific actions on specific queues. For example, in case your function only polls from the queue, grant it permission to read messages, but not to send new messages. A function execution role defines which actions your function is allowed to perform on other resources. A queue access policy defines the principals that can access this queue, and the actions that are allowed.

Use server-side encryption (SSE) to store sensitive data in encrypted SQS queues. With SSE, your messages are always stored in encrypted form, and SQS only decrypts them for sending to an authorized consumer. SSE protects the contents of messages in queues using SQS-managed encryption keys (SSE-SQS) or keys managed in the AWS Key Management Service (SSE-KMS).

Conclusion

The improved Lambda SQS event source polling scale-up capability enables up to five times faster scale-up performance for spiky event-driven workloads using SQS queues, at no additional cost. This improvement offers customers the benefit of faster processing during spikes of messages in SQS queues, while continuing to offer the flexibility to limit the maximum concurrent invokes by SQS as an event source.

For more serverless learning resources, visit Serverless Land.

Orchestrating dependent file uploads with AWS Step Functions

2023-11-02 Benjamin Smith

Post Syndicated from Benjamin Smith original https://aws.amazon.com/blogs/compute/orchestrating-dependent-file-uploads-with-aws-step-functions/

This post is written by Nelson Assis, Enterprise Support Lead, Serverless and Jevon Liburd, Technical Account Manager, Serverless

Amazon S3 is an object storage service that many customers use for file storage. With the use of Amazon S3 Event Notifications or Amazon EventBridge customers can create workloads with event-driven architecture (EDA). This architecture responds to events produced when changes occur to objects in S3 buckets.

EDA involves asynchronous communication between system components. This serves to decouple the components allowing each component to be autonomous.

Some scenarios may introduce coupling in the architecture due to dependency between events. This blog post presents a common example of this coupling and how it can be handled using AWS Step Functions.

Overview

In this example, an organization has two distributed autonomous teams, the Sales team and the Warehouse team. Each team is responsible for uploading a monthly data file to an S3 bucket so it can be processed.

The files generate events when they are uploaded, initiating downstream processes. The processing of the Warehouse file cleans the data and joins it with data from the Shipping team. The processing of the Sales file correlates the data with the combined Warehouse and Shipping data. This enables analysts to perform forecasting and gain other insights.

For this correlation to happen, the Warehouse file must be processed before the Sales file. As the two teams are autonomous, there is no coordination among the teams. This means that the files can be uploaded at any time with no assurance that the Warehouse file is processed before the Sales file.

For scenarios like these, the Aggregator pattern can be used. The pattern collects and stores the events, and triggers a new event based on the combined events. In the described scenario, the combined events are the processed Warehouse file and the uploaded Sales file.

The requirements of the aggregator pattern are:

Correlation – A way to group the related events. This is fulfilled by a unique identifier in the file name.
Event aggregator – A stateful store for the events.
Completion check and trigger – A condition when the combined events have been received and a way to publish the resulting event.

Architecture overview

The architecture uses the following AWS services:

Amazon DynamoDB as the event aggregator.
Step Functions to orchestrate the workflow.
AWS Lambda to parse the file name and extract the correlation identifier.
AWS Serverless Application Model (AWS SAM) for infrastructure as code and deployment.

File upload: The Sales and Warehouse teams upload their respective files to S3.
EventBridge: The ObjectCreated event is sent to EventBridge where there is a rule with a target of the main workflow.
Main state machine: This state machine orchestrates the aggregator operations and the processing of the files. It encapsulates the workflows for each file to separate the aggregator logic from the files’ workflow logic.
File parser and correlation: The business logic to identify the file and its type is run in this Lambda function.
Stateful store: A DynamoDB table stores information about the file such as the name, type, and processing status. The state machine reads from and writes to the DynamoDB table. Task tokens are also stored in this table.
File processing: Depending on the file type and any pre-conditions, state machines corresponding to the file type are run. These state machines contain the logic to process the specific file.
Task Token & Callback: The task token is generated when the dependent file tries to be processed before the independent file. The Step Functions “Wait for a Callback” pattern continues the execution of the dependent file after the independent file is processed.

Walkthrough

You need the following prerequisites:

AWS CLI and AWS SAM CLI installed.
An AWS account.
Sufficient permissions to manage the AWS resources.
Git installed.

To deploy the example, follow the instructions in the GitHub repo.

This walkthrough shows what happens if the dependent file (Sales file) is uploaded before the independent one (Warehouse file).

The workflow starts with the uploading of the Sales file to the dedicated Sales S3 bucket. The example uses separate S3 buckets for the two files as it assumes that the Sales and Warehouse teams are distributed and autonomous. You can find sample files in the code repository.

Uploading the file to S3 sends an event to EventBridge, which the aggregator state machine acts on. The event pattern used in the EventBridge rule is:

{
  "detail-type": ["Object Created"],
  "source": ["aws.s3"],
  "detail": {
    "bucket": {
      "name": ["sales-mfu-eda-09092023", "warehouse-mfu-eda-09092023"]
    },
    "reason": ["PutObject"]
  }
}

The aggregator state machine starts by invoking the file parser Lambda function. This function parses the file type and uses the identifier to correlate the files. In this example, the name of the file contains the file type and the correlation identifier (the year_month). To use other ways of representing the file type and correlation identifier, you can modify this function to parse that information.
The next step in the state machine inserts a record for the event in the event aggregator DynamoDB table. The table has a composite primary key with the correlation identifier as the partition key and the file type as the sort key. The processing status of the file is tracked to give feedback on the state of the workflow.
Based on the file type, the state machine determines which branch to follow. In the example, the Sales branch is run. The state machine tries to get the status of the (dependent) Warehouse file from DynamoDB using the correlation identifier. Using the result of this query, the state machine determines if the corresponding Warehouse file has already been processed.
Since the Warehouse file is not processed yet, the waitForTaskToken integration pattern is used. The state machine waits at this step and creates a task token, which the external services use to trigger the state machine to continue its execution. The Sales record in the DynamoDB table is updated with the Task Token.
Navigate to the S3 console and upload the sample Warehouse file to the Warehouse S3 bucket. This invokes a new instance of the Step Functions workflow, which flows through the other branch after the file type choice step. In this branch, the Warehouse state machine is run and the processing status of the file is updated in DynamoDB.

When the status of the Warehouse file is changed to “Completed”, the Warehouse state machine checks DynamoDB for a pending Sales file. If there is one, it retrieves the task token and calls the SendTaskSuccess method. This triggers the Sales state machine, which is in a waiting state to continue. The Sales state machine is started and the processing status is updated.

Conclusion

This blog post shows how to handle file dependencies in event driven architectures. You can customize the sample provided in the code repository for your own use case.

This solution is specific to file dependencies in event driven architectures. For more information on solving event dependencies and aggregators read the blog post: Moving to event-driven architectures with serverless event aggregators.

To learn more about event driven architectures, visit the event driven architecture section on Serverless Land.

GoDaddy benchmarking results in up to 24% better price-performance for their Spark workloads with AWS Graviton2 on Amazon EMR Serverless

2023-11-02 Mukul Sharma

Post Syndicated from Mukul Sharma original https://aws.amazon.com/blogs/big-data/godaddy-benchmarking-results-in-up-to-24-better-price-performance-for-their-spark-workloads-with-aws-graviton2-on-amazon-emr-serverless/

This is a guest post co-written with Mukul Sharma, Software Development Engineer, and Ozcan IIikhan, Director of Engineering from GoDaddy.

GoDaddy empowers everyday entrepreneurs by providing all the help and tools to succeed online. With more than 22 million customers worldwide, GoDaddy is the place people come to name their ideas, build a professional website, attract customers, and manage their work.

GoDaddy is a data-driven company, and getting meaningful insights from data helps us drive business decisions to delight our customers. At GoDaddy, we embarked on a journey to uncover the efficiency promises of AWS Graviton2 on Amazon EMR Serverless as part of our long-term vision for cost-effective intelligent computing.

In this post, we share the methodology and results of our benchmarking exercise comparing the cost-effectiveness of EMR Serverless on the arm64 (Graviton2) architecture against the traditional x86_64 architecture. EMR Serverless on Graviton2 demonstrated an advantage in cost-effectiveness, resulting in significant savings in total run costs. We achieved 23.85% improvement in price-performance for sample production Spark workloads—an outcome that holds tremendous potential for businesses striving to maximize their computing efficiency.

Solution overview

GoDaddy’s intelligent compute platform envisions simplification of compute operations for all personas, without limiting power users, to ensure out-of-box cost and performance optimization for data and ML workloads. As a part of this vision, GoDaddy’s Data & ML Platform team plans to use EMR Serverless as one of the compute solutions under the hood.

The following diagram shows a high-level illustration of the intelligent compute platform vision.

Benchmarking EMR Serverless for GoDaddy

EMR Serverless is a serverless option in Amazon EMR that eliminates the complexities of configuring, managing, and scaling clusters when running big data frameworks like Apache Spark and Apache Hive. With EMR Serverless, businesses can enjoy numerous benefits, including cost-effectiveness, faster provisioning, simplified developer experience, and improved resilience to Availability Zone failures.

At GoDaddy, we embarked on a comprehensive study to benchmark EMR Serverless using real production workflows at GoDaddy. The purpose of the study was to evaluate the performance and efficiency of EMR Serverless and develop a well-informed adoption plan. The results of the study have been extremely promising, showcasing the potential of EMR Serverless for our workloads.

Having achieved compelling results in favor of EMR Serverless for our workloads, our attention turned to evaluating the utilization of the Graviton2 (arm64) architecture on EMR Serverless. In this post, we focus on comparing the performance of Graviton2 (arm64) with the x86_64 architecture on EMR Serverless. By conducting this apples-to-apples comparative analysis, we aim to gain valuable insights into the benefits and considerations of using Graviton2 for our big data workloads.

By using EMR Serverless and exploring the performance of Graviton2, GoDaddy aims to optimize their big data workflows and make informed decisions regarding the most suitable architecture for their specific needs. The combination of EMR Serverless and Graviton2 presents an exciting opportunity to enhance the data processing capabilities and drive efficiency in our operations.

AWS Graviton2

The Graviton2 processors are specifically designed by AWS, utilizing powerful 64-bit Arm Neoverse cores. This custom-built architecture provides a remarkable boost in price-performance for various cloud workloads.

In terms of cost, Graviton2 offers an appealing advantage. As indicated in the following table, the pricing for Graviton2 is 20% lower compared to the x86 architecture option.

	x86_64	arm64 (Graviton2)
per vCPU per hour	$0.052624	$0.042094
per GB per hour	$0.0057785	$0.004628
per storage GB per hour*	$0.000111

*Ephemeral storage: 20 GB of ephemeral storage is available for all workers by default—you pay only for any additional storage that you configure per worker.

For specific pricing details and current information, refer to Amazon EMR pricing.

AWS benchmark

The AWS team performed benchmark tests on Spark workloads with Graviton2 on EMR Serverless using the TPC-DS 3 TB scale performance benchmarks. The summary of their analysis are as follows:

Graviton2 on EMR Serverless demonstrated an average improvement of 10% for Spark workloads in terms of runtime. This indicates that the runtime for Spark-based tasks was reduced by approximately 10% when utilizing Graviton2.
Although the majority of queries showcased improved performance, a small subset of queries experienced a regression of up to 7% on Graviton2. These specific queries showed a slight decrease in performance compared to the x86 architecture option.
In addition to the performance analysis, the AWS team considered the cost factor. Graviton2 is offered at a 20% lower cost than the x86 architecture option. Taking this cost advantage into account, the AWS benchmark set yielded an overall 27% better price-performance for workloads. This means that by using Graviton2, users can achieve a 27% improvement in performance per unit of cost compared to the x86 architecture option.

These findings highlight the significant benefits of using Graviton2 on EMR Serverless for Spark workloads, with improved performance and cost-efficiency. It showcases the potential of Graviton2 in delivering enhanced price-performance ratios, making it an attractive choice for organizations seeking to optimize their big data workloads.

GoDaddy benchmark

During our initial experimentation, we observed that arm64 on EMR Serverless consistently outperformed or performed on par with x86_64. One of the jobs showed a 7.51% increase in resource usage on arm64 compared to x86_64, but due to the lower price of arm64, it still resulted in a 13.48% cost reduction. In another instance, we achieved an impressive 43.7% reduction in run cost, attributed to both the lower price and reduced resource utilization. Overall, our initial tests indicated that arm64 on EMR Serverless delivered superior price-performance compared to x86_64. These promising findings motivated us to conduct a more comprehensive and rigorous study.

Benchmark results

To gain a deeper understanding of the value of Graviton2 on EMR Serverless, we conducted our study using real-life production workloads from GoDaddy, which are scheduled to run at a daily cadence. Without any exceptions, EMR Serverless on arm64 (Graviton2) is significantly more cost-effective compared to the same jobs run on EMR Serverless on the x86_64 architecture. In fact, we recorded an impressive 23.85% improvement in price-performance across the sample GoDaddy jobs using Graviton2.

Like the AWS benchmarks, we observed slight regressions of less than 5% in the total runtime of some jobs. However, given that these jobs will be migrated from Amazon EMR on EC2 to EMR Serverless, the overall total runtime will still be shorter due to the minimal provisioning time in EMR Serverless. Additionally, across all jobs, we observed an average speed up of 2.1% in addition to the cost savings achieved.

These benchmarking results provide compelling evidence of the value and effectiveness of Graviton2 on EMR Serverless. The combination of improved price-performance, shorter runtimes, and overall cost savings makes Graviton2 a highly attractive option for optimizing big data workloads.

Benchmarking methodology

As an extension of a larger benchmarking EMR Serverless for GoDaddy study, where we divided Spark jobs into brackets based on total runtime (quick-run, medium-run, long-run), we measured effect of architecture (arm64 vs. x86_64) on total cost and total runtime. All other parameters were kept the same to achieve an apples-to-apples comparison.

The team followed these steps:

Prepare the data and environment.
Choose two random production jobs from each job bracket.
Make necessary changes to avoid inference with actual production outputs.
Run tests to execute scripts over multiple iterations to collect accurate and consistent data points.
Validate input and output datasets, partitions, and row counts to ensure identical data processing.
Gather relevant metrics from the tests.
Analyze results to draw insights and conclusions.

The following table shows the summary of an example Spark job.

Metric	EMR Serverless (Average) – X86_64	EMR Serverless (Average) – Graviton	X86_64 vs Graviton (% Difference)
Total Run Cost	$2.76	$1.85	32.97%
Total Runtime (hh:mm:ss)	00:41:31	00:34:32	16.82%
EMR Release Label	emr-6.9.0
Job Type	Spark
Spark Version	Spark 3.3.0
Hadoop Distribution	Amazon 3.3.3
Hive/HCatalog Version	Hive 3.1.3, HCatalog 3.1.3

Summary of results

The following table presents a comparison of job performance between EMR Serverless on arm64 (Graviton2) and EMR Serverless on x86_64. For each architecture, every job was run at least three times to obtain the accurate average cost and runtime.

Job	Average x86_64 Cost	Average arm64 Cost	Average x86_64 Runtime (hh:mm:ss)	Average arm64 Runtime (hh:mm:ss)	Average Cost Savings %	Average Performance Gain %
1	$1.64	$1.25	00:08:43	00:09:01	23.89%	-3.24%
2	$10.00	$8.69	00:27:55	00:28:25	13.07%	-1.79%
3	$29.66	$24.15	00:50:49	00:53:17	18.56%	-4.85%
4	$34.42	$25.80	01:20:02	01:24:54	25.04%	-6.08%
5	$2.76	$1.85	00:41:31	00:34:32	32.97%	16.82%
6	$34.07	$24.00	00:57:58	00:51:09	29.57%	11.76%
Average					23.85%	2.10%

Note that the improvement calculations are based on higher-precision results for more accuracy.

Conclusion

Based on this study, GoDaddy observed a significant 23.85% improvement in price-performance for sample production Spark jobs utilizing the arm64 architecture compared to the x86_64 architecture. These compelling results have led us to strongly recommend internal teams to use arm64 (Graviton2) on EMR Serverless, except in cases where there are compatibility issues with third-party packages and libraries. By adopting an arm64 architecture, organizations can achieve enhanced cost-effectiveness and performance for their workloads, contributing to more efficient data processing and analytics.

About the Authors

Mukul Sharma is a Software Development Engineer on Data & Analytics (DnA) organization at GoDaddy. He is a polyglot programmer with experience in a wide array of technologies to rapidly deliver scalable solutions. He enjoys singing karaoke, playing various board games, and working on personal programming projects in his spare time.

Ozcan Ilikhan is a Director of Engineering on Data & Analytics (DnA) organization at GoDaddy. He is passionate about solving customer problems and increasing efficiency using data and ML/AI. In his spare time, he loves reading, hiking, gardening, and working on DIY projects.

Harsh Vardhan Singh Gaur is an AWS Solutions Architect, specializing in analytics. He has over 6 years of experience working in the field of big data and data science. He is passionate about helping customers adopt best practices and discover insights from their data.

Ramesh Kumar Venkatraman is a Senior Solutions Architect at AWS who is passionate about containers and databases. He works with AWS customers to design, deploy, and manage their AWS workloads and architectures. In his spare time, he loves to play with his two kids and follows cricket.

Transforming transactions: Streamlining PCI compliance using AWS serverless architecture

2023-10-31 Abdul Javid

Post Syndicated from Abdul Javid original https://aws.amazon.com/blogs/security/transforming-transactions-streamlining-pci-compliance-using-aws-serverless-architecture/

Compliance with the Payment Card Industry Data Security Standard (PCI DSS) is critical for organizations that handle cardholder data. Achieving and maintaining PCI DSS compliance can be a complex and challenging endeavor. Serverless technology has transformed application development, offering agility, performance, cost, and security.

In this blog post, we examine the benefits of using AWS serverless services and highlight how you can use them to help align with your PCI DSS compliance responsibilities. You can remove additional undifferentiated compliance heavy lifting by building modern applications with abstracted AWS services. We review an example payment application and workflow that uses AWS serverless services and showcases the potential reduction in effort and responsibility that a serverless architecture could provide to help align with your compliance requirements. We present the review through the lens of a merchant that has an ecommerce website and include key topics such as access control, data encryption, monitoring, and auditing—all within the context of the example payment application. We don’t discuss additional service provider requirements from the PCI DSS in this post.

This example will help you navigate the intricate landscape of PCI DSS compliance. This can help you focus on building robust and secure payment solutions without getting lost in the complexities of compliance. This can also help reduce your compliance burden and empower you to develop your own secure, scalable applications. Join us in this journey as we explore how AWS serverless services can help you meet your PCI DSS compliance objectives.

Disclaimer

This document is provided for the purposes of information only; it is not legal advice, and should not be relied on as legal advice. Customers are responsible for making their own independent assessment of the information in this document. This document: (a) is for informational purposes only, (b) represents current AWS product offerings and practices, which are subject to change without notice, and (c) does not create any commitments or assurances from AWS and its affiliates, suppliers or licensors. AWS products or services are provided “as is” without warranties, representations, or conditions of any kind, whether express or implied. The responsibilities and liabilities of AWS to its customers are controlled by AWS agreements, and this document is not part of, nor does it modify, any agreement between AWS and its customers.

AWS encourages its customers to obtain appropriate advice on their implementation of privacy and data protection environments, and more generally, applicable laws and other obligations relevant to their business.

PCI DSS v4.0 and serverless

In April 2022, the Payment Card Industry Security Standards Council (PCI SSC) updated the security payment standard to “address emerging threats and technologies and enable innovative methods to combat new threats.” Two of the high-level goals of these updates are enhancing validation methods and procedures and promoting security as a continuous process. Adopting serverless architectures can help meet some of the new and updated requirements in version 4.0, such as enhanced software and encryption inventories. If a customer has access to change a configuration, it’s the customer’s responsibility to verify that the configuration meets PCI DSS requirements. There are more than 20 PCI DSS requirements applicable to Amazon Elastic Compute Cloud (Amazon EC2). To fulfill these requirements, customer organizations must implement controls such as file integrity monitoring, operating system level access management, system logging, and asset inventories. Using AWS abstracted services in this scenario can remove undifferentiated heavy lifting from your environment. With abstracted AWS services, because there is no operating system to manage, AWS becomes responsible for maintaining consistent time settings for an abstracted service to meet Requirement 10.6. This will also shift your compliance focus more towards your application code and data.

This makes more of your PCI DSS responsibility addressable through the AWS PCI DSS Attestation of Compliance (AOC) and Responsibility Summary. This attestation package is available to AWS customers through AWS Artifact.

Reduction in compliance burden

You can use three common architectural patterns within AWS to design payment applications and meet PCI DSS requirements: infrastructure, containerized, and abstracted. We look into EC2 instance-based architecture (infrastructure or containerized patterns) and modernized architectures using serverless services (abstracted patterns). While both approaches can help align with PCI DSS requirements, there are notable differences in how they handle certain elements. EC2 instances provide more control and flexibility over the underlying infrastructure and operating system, assisting you in customizing security measures based on your organization’s operational and security requirements. However, this also means that you bear more responsibility for configuring and maintaining security controls applicable to the operating systems, such as network security controls, patching, file integrity monitoring, and vulnerability scanning.

On the other hand, serverless architectures similar to the preceding example can reduce much of the infrastructure management requirements. This can relieve you, the application owner or cloud service consumer, of the burden of configuring and securing those underlying virtual servers. This can streamline meeting certain PCI requirements, such as file integrity monitoring, patch management, and vulnerability management, because AWS handles these responsibilities.

Using serverless architecture on AWS can significantly reduce the PCI compliance burden. Approximately 43 percent of the overall PCI compliance requirements, encompassing both technical and non-technical tests, are addressed by the AWS PCI DSS Attestation of Compliance.

Customer responsible
52%

AWS responsible
43%

N/A
5%

The following table provides an analysis of each PCI DSS requirement against the serverless architecture in Figure 1, which shows a sample payment application workflow. You must evaluate your own use and secure configuration of AWS workload and architectures for a successful audit.

PCI DSS 4.0 requirements	Test cases	Customer responsible	AWS responsible	N/A
Requirement 1: Install and maintain network security controls	35	13	22	0
Requirement 2: Apply secure configurations to all system components	27	16	11	0
Requirement 3: Protect stored account data	55	24	29	2
Requirement 4: Protect cardholder data with strong cryptography during transmission over open, public networks	12	7	5	0
Requirement 5: Protect all systems and networks from malicious software	25	4	21	0
Requirement 6: Develop and maintain secure systems and software	35	31	4	0
Requirement 7: Restrict access to system components and cardholder data by business need-to-know	22	19	3	0
Requirement 8: Identify users and authenticate access to system components	52	43	6	3
Requirement 9: Restrict physical access to cardholder data	56	3	53	0
Requirement 10: Log and monitor all access to system components and cardholder data	38	17	19	2
Requirement 11: Test security of systems and networks regularly	51	22	23	6
Requirement 12: Support information security with organizational policies	56	44	2	10
Total	464	243	198	23
Percentage		52%	43%	5%

Note: The preceding table is based on the example reference architecture that follows. The actual extent of PCI DSS requirements reduction can vary significantly depending on your cardholder data environment (CDE) scope, implementation, and configurations.

Sample payment application and workflow

This example serverless payment application and workflow in Figure 1 consists of several interconnected steps, each using different AWS services. The steps are listed in the following text and include brief descriptions. They cover two use cases within this example application — consumers making a payment and a business analyst generating a report.

The example outlines a basic serverless payment application workflow using AWS serverless services. However, it’s important to note that the actual implementation and behavior of the workflow may vary based on specific configurations, dependencies, and external factors. The example serves as a general guide and may require adjustments to suit the unique requirements of your application or infrastructure.

Several factors, including but not limited to, AWS service configurations, network settings, security policies, and third-party integrations, can influence the behavior of the system. Before deploying a similar solution in a production environment, we recommend thoroughly reviewing and adapting the example to align with your specific use case and requirements.

Keep in mind that AWS services and features may evolve over time, and new updates or changes may impact the behavior of the components described in this example. Regularly consult the AWS documentation and ensure that your configurations adhere to best practices and compliance standards.

This example is intended to provide a starting point and should be considered as a reference rather than an exhaustive solution. Always conduct thorough testing and validation in your specific environment to ensure the desired functionality and security.

Figure 1: Serverless payment architecture and workflow

Use case 1: Consumers make a payment
1. Consumers visit the e-commerce payment page to make a payment.
2. The request is routed to the payment application’s domain using Amazon Route 53, which acts as a DNS service.
3. The payment page is protected by AWS WAF to inspect the initial incoming request for any malicious patterns, web-based attacks (such as cross-site scripting (XSS) attacks), and unwanted bots.
4. An HTTPS GET request (over TLS) is sent to the public target IP. Amazon CloudFront, a content delivery network (CDN), acts as a front-end proxy and caches and fetches static content from an Amazon Simple Storage Service (Amazon S3) bucket.
5. AWS WAF inspects the incoming request for any malicious patterns, if the request is blocked, the request doesn’t return static content from the S3 bucket.
6. User authentication and authorization are handled by Amazon Cognito, providing a secure login and scalable customer identity and access management system (CIAM)
7. AWS WAF processes the request to protect against web exploits, then Amazon API Gateway forwards it to the payment application API endpoint.
8. API Gateway launches AWS Lambda functions to handle payment requests. AWS Step Functions state machine oversees the entire process, directing the running of multiple Lambda functions to communicate with the payment processor, initiate the payment transaction, and process the response.
9. The cardholder data (CHD) is temporarily cached in Amazon DynamoDB for troubleshooting and retry attempts in the event of transaction failures.
10. A Lambda function validates the transaction details and performs necessary checks against the data stored in DynamoDB. A web notification is sent to the consumer for any invalid data.
11. A Lambda function calculates the transaction fees.
12. A Lambda function authenticates the transaction and initiates the payment transaction with the third-party payment provider.
13. A Lambda function is initiated when a payment transaction with the third-party payment provider is completed. It receives the transaction status from the provider and performs multiple actions.
14. Consumers receive real-time notifications through a web browser and email. The notifications are initiated by a step function, such as order confirmations or payment receipts, and can be integrated with external payment processors through an Amazon Simple Notification Service (Amazon SNS) Amazon Simple Email Service (Amazon SES) web hook.
15. A separate Lambda function clears the DynamoDB cache.
16. The Lambda function makes entries into the Amazon Simple Queue Service (Amazon SQS) dead-letter queue for failed transactions to retry at a later time.
Use case 2: An admin or analyst generates the report for non-PCI data
1. An admin accesses the web-based reporting dashboard using their browser to generate a report.
2. The request is routed to AWS WAF to verify the source that initiated the request.
3. An HTTPS GET request (over TLS) is sent to the public target IP. CloudFront fetches static content from an S3 bucket.
4. AWS WAF inspects incoming requests for any malicious patterns, if the request is blocked, the request doesn’t return static content from the S3 bucket. The validated traffic is sent to Amazon S3 to retrieve the reporting page.
5. The backend requests of the reporting page pass through AWS WAF again to provide protection against common web exploits before being forwarded to the reporting API endpoint through API Gateway.
6. API Gateway launches a Lambda function for report generation. The Lambda function retrieves data from DynamoDB storage for the reporting mechanism.
7. The AWS Security Token Service (AWS STS) issues temporary credentials to the Lambda service in the non-PCI serverless account, allowing it to launch the Lambda function in the PCI serverless account. The Lambda function retrieves non-PCI data and writes it into DynamoDB.
8. The Lambda function fetches the non-PCI data based on the report criteria from the DynamoDB table from the same account.

Additional AWS security and governance services that would be implemented throughout the architecture are shown in Figure 1, Label-25. For example, Amazon CloudWatch monitors and alerts on all the Lambda functions within the environment.

Label-26 demonstrates frameworks that can be used to build the serverless applications.

Scoping and requirements

Now that we’ve established the reference architecture and workflow, lets delve into how it aligns with PCI DSS scope and requirements.

PCI scoping

Serverless services are inherently segmented by AWS, but they can be used within the context of an AWS account hierarchy to provide various levels of isolation as described in the reference architecture example.

Segregating PCI data and non-PCI data into separate AWS accounts can help in de-scoping non-PCI environments and reducing the complexity and audit requirements for components that don’t handle cardholder data.

PCI serverless production account

This AWS account is dedicated to handling PCI data and applications that directly process, transmit, or store cardholder data.
Services such as Amazon Cognito, DynamoDB, API Gateway, CloudFront, Amazon SNS, Amazon SES, Amazon SQS, and Step Functions are provisioned in this account to support the PCI data workflow.
Security controls, logging, monitoring, and access controls in this account are specifically designed to meet PCI DSS requirements.

Non-PCI serverless production account

This separate AWS account is used to host applications that don’t handle PCI data.
Since this account doesn’t handle cardholder data, the scope of PCI DSS compliance is reduced, simplifying the compliance process.

Note: You can use AWS Organizations to centrally manage multiple AWS accounts.

AWS IAM Identity Center (successor to AWS Single Sign-On) is used to manage user access to each account and is integrated with your existing identify provider. This helps to ensure you’re meeting PCI requirements on identity, access control of card holder data, and environment.

Now, let’s look at the PCI DSS requirements that this architectural pattern can help address.

Requirement 1: Install and maintain network security controls

Network security controls are limited to AWS Identity and Access Management (IAM) and application permissions because there is no customer controlled or defined network. VPC-centric requirements aren’t applicable because there is no VPC. The configuration settings for serverless services can be covered under Requirement 6 to for secure configuration standards. This supports compliance with Requirements 1.2 and 1.3.

Requirement 2: Apply secure configurations to all system components

AWS services are single function by default and exist with only the necessary functionality enabled for the functioning of that service. This supports compliance with much of Requirement 2.2.
Access to AWS services is considered non-console and only accessible through HTTPS through the service API. This supports compliance with Requirement 2.2.7.
The wireless requirements under Requirement 2.3 are not applicable, because wireless environments don’t exist in AWS environments.

Requirement 3: Protect stored account data

AWS is responsible for destruction of account data configured for deletion based on DynamoDB Time to Live (TTL) values. This supports compliance with Requirement 3.2.
DynamoDB and Amazon S3 offer secure storage of account data, encryption by default in transit and at rest, and integration with AWS Key Management Service (AWS KMS). This supports compliance with Requirements 3.5 and 4.2.
AWS is responsible for the generation, distribution, storage, rotation, destruction, and overall protection of encryption keys within AWS KMS. This supports compliance with Requirements 3.6 and 3.7.
Manual cleartext cryptographic keys aren’t available in this solution, Requirement 3.7.6 is not applicable.

Requirement 4: Protect cardholder data with strong cryptography during transmission over open, public networks

AWS Certificate Manager (ACM) integrates with API Gateway and enables the use of trusted certificates and HTTPS (TLS) for secure communication between clients and the API. This supports compliance with Requirement 4.2.
Requirement 4.2.1.2 is not applicable because there are no wireless technologies in use in this solution. Customers are responsible for ensuring strong cryptography exists for authentication and transmission over other wireless networks they manage outside of AWS.
Requirement 4.2.2 is not applicable because no end-user technologies exist in this solution. Customers are responsible for ensuring the use of strong cryptography if primary account numbers (PAN) are sent through end-user messaging technologies in other environments.

Requirement 5: Protect a ll systems and networks from malicious software

There are no customer-managed compute resources in this example payment environment, Requirements 5.2 and 5.3 are the responsibility of AWS.

Requirement 6: Develop and maintain secure systems and software

Amazon Inspector now supports Lambda functions, adding continual, automated vulnerability assessments for serverless compute. This supports compliance with Requirement 6.2.
Amazon Inspector helps identify vulnerabilities and security weaknesses in the payment application’s code, dependencies, and configuration. This supports compliance with Requirement 6.3.
AWS WAF is designed to protect applications from common attacks, such as SQL injections, cross-site scripting, and other web exploits. AWS WAF can filter and block malicious traffic before it reaches the application. This supports compliance with Requirement 6.4.2.

Requirement 7: Restrict access to system components and cardholder data by business need to know

IAM and Amazon Cognito allow for fine-grained role- and job-based permissions and access control. Customers can use these capabilities to configure access following the principles of least privilege and need-to-know. IAM and Cognito support the use of strong identification, authentication, authorization, and multi-factor authentication (MFA). This supports compliance with much of Requirement 7.

Requirement 8: Identify users and authenticate access to system components

IAM and Amazon Cognito also support compliance with much of Requirement 8.
Some of the controls in this requirement are usually met by the identity provider for internal access to the cardholder data environment (CDE).

Requirement 9: Restrict physical access to cardholder data

AWS is responsible for the destruction of data in DynamoDB based on the customer configuration of content TTL values for Requirement 9.4.7. Customers are responsible for ensuring their database instance is configured for appropriate removal of data by enabling TTL on DDB attributes.
Requirement 9 is otherwise not applicable for this serverless example environment because there are no physical media, electronic media not already addressed under Requirement 3.2, or hard-copy materials with cardholder data. AWS is responsible for the physical infrastructure under the Shared Responsibility Model.

Requirement 10: Log and monitor all access to system components and cardholder data

AWS CloudTrail provides detailed logs of API activity for auditing and monitoring purposes. This supports compliance with Requirement 10.2 and contains all of the events and data elements listed.
CloudWatch can be used for monitoring and alerting on system events and performance metrics. This supports compliance with Requirement 10.4.
AWS Security Hub provides a comprehensive view of security alerts and compliance status, consolidating findings from various security services, which helps in ongoing security monitoring and testing. Customers must enable PCI DSS security standard, which supports compliance with Requirement 10.4.2.
AWS is responsible for maintaining accurate system time for AWS services. In this example, there are no compute resources for which customers can configure time. Requirement 10.6 is addressable through the AWS Attestation of Compliance and Responsibility Summary available in AWS Artifact.

Requirement 11: Regularly test security systems and processes

Testing for rogue wireless activity within the AWS-based CDE is the responsibility of AWS. AWS is responsible for the management of the physical infrastructure under Requirement 11.2. Customers are still responsible for wireless testing for their environments outside of AWS, such as where administrative workstations exist.
AWS is responsible for internal vulnerability testing of AWS services, and supports compliance with Requirement 11.3.1.
Amazon GuardDuty, a threat detection service that continuously monitors for malicious activity and unauthorized access, providing continuous security monitoring. This supports the IDS requirements under Requirement 11.5.1, and covers the entire AWS-based CDE.
AWS Config allows customers to catalog, monitor and manage configuration changes for their AWS resources. This supports compliance with Requirement 11.5.2.
Customers can use AWS Config to monitor the configuration of the S3 bucket hosting the static website. This supports compliance with Requirement 11.6.1.

Requirement 12: Support information security with organizational policies and programs

Customers can download the AWS AOC and Responsibility Summary package from Artifact to support Requirement 12.8.5 and the identification of which PCI DSS requirements are managed by the third-party service provider (TSPS) and which by the customer.

Conclusion

Using AWS serverless services when developing your payment application can significantly help reduce the number of PCI DSS requirements you need to meet by yourself. By offloading infrastructure management to AWS and using serverless services such as Lambda, API Gateway, DynamoDB, Amazon S3, and others, you can benefit from built-in security features and help align with your PCI DSS compliance requirements.

Contact us to help design an architecture that works for your organization. AWS Security Assurance Services is a Payment Card Industry-Qualified Security Assessor company (PCI-QSAC) and HITRUST External Assessor firm. We are a team of industry-certified assessors who help you to achieve, maintain, and automate compliance in the cloud by tying together applicable audit standards to AWS service-specific features and functionality. We help you build on frameworks such as PCI DSS, HITRUST CSF, NIST, SOC 2, HIPAA, ISO 27001, GDPR, and CCPA.

More information on how to build applications using AWS serverless technologies can be found at Serverless on AWS.

Want more AWS Security news? Follow us on Twitter.

If you have feedback about this post, submit comments in the Comments section below. If you have questions about this post, start a new thread on the Serverless re:Post, Security, Identity, & Compliance re:Post or contact AWS Support.

Sending and receiving webhooks on AWS: Innovate with event notifications

2023-10-30 James Beswick

Post Syndicated from James Beswick original https://aws.amazon.com/blogs/compute/sending-and-receiving-webhooks-on-aws-innovate-with-event-notifications/

This post is written by Daniel Wirjo, Solutions Architect, and Justin Plock, Principal Solutions Architect.

Commonly known as reverse APIs or push APIs, webhooks provide a way for applications to integrate to each other and communicate in near real-time. It enables integration for business and system events.

Whether you’re building a software as a service (SaaS) application integrating with your customer workflows, or transaction notifications from a vendor, webhooks play a critical role in unlocking innovation, enhancing user experience, and streamlining operations.

This post explains how to build with webhooks on AWS and covers two scenarios:

Webhooks Provider: A SaaS application that sends webhooks to an external API.
Webhooks Consumer: An API that receives webhooks with capacity to handle large payloads.

It includes high-level reference architectures with considerations, best practices and code sample to guide your implementation.

Sending webhooks

To send webhooks, you generate events, and deliver them to third-party APIs. These events facilitate updates, workflows, and actions in the third-party system. For example, a payments platform (provider) can send notifications for payment statuses, allowing ecommerce stores (consumers) to ship goods upon confirmation.

AWS reference architecture for a webhook provider

The architecture consists of two services:

Webhook delivery: An application that delivers webhooks to an external endpoint specified by the consumer.
Subscription management: A management API enabling the consumer to manage their configuration, including specifying endpoints for delivery, and which events for subscription.

Considerations and best practices for sending webhooks

When building an application to send webhooks, consider the following factors:

Event generation: Consider how you generate events. This example uses Amazon DynamoDB as the data source. Events are generated by change data capture for DynamoDB Streams and sent to Amazon EventBridge Pipes. You then simplify the DynamoDB response format by using an input transformer.

With EventBridge, you send events in near real time. If events are not time-sensitive, you can send multiple events in a batch. This can be done by polling for new events at a specified frequency using EventBridge Scheduler. To generate events from other data sources, consider similar approaches with Amazon Simple Storage Service (S3) Event Notifications or Amazon Kinesis.

Filtering: EventBridge Pipes support filtering by matching event patterns, before the event is routed to the target destination. For example, you can filter for events in relation to status update operations in the payments DynamoDB table to the relevant subscriber API endpoint.

Delivery: EventBridge API Destinations deliver events outside of AWS using REST API calls. To protect the external endpoint from surges in traffic, you set an invocation rate limit. In addition, retries with exponential backoff are handled automatically depending on the error. An Amazon Simple Queue Service (SQS) dead-letter queue retains messages that cannot be delivered. These can provide scalable and resilient delivery.

Payload Structure: Consider how consumers process event payloads. This example uses an input transformer to create a structured payload, aligned to the CloudEvents specification. CloudEvents provides an industry standard format and common payload structure, with developer tools and SDKs for consumers.

Payload Size: For fast and reliable delivery, keep payload size to a minimum. Consider delivering only necessary details, such as identifiers and status. For additional information, you can provide consumers with a separate API. Consumers can then separately call this API to retrieve the additional information.

Security and Authorization: To deliver events securely, you establish a connection using an authorization method such as OAuth. Under the hood, the connection stores the credentials in AWS Secrets Manager, which securely encrypts the credentials.

Subscription Management: Consider how consumers can manage their subscription, such as specifying HTTPS endpoints and event types to subscribe. DynamoDB stores this configuration. Amazon API Gateway, Amazon Cognito, and AWS Lambda provide a management API for operations.

Costs: In practice, sending webhooks incurs cost, which may become significant as you grow and generate more events. Consider implementing usage policies, quotas, and allowing consumers to subscribe only to the event types that they need.

Monetization: Consider billing consumers based on their usage volume or tier. For example, you can offer a free tier to provide a low-friction access to webhooks, but only up to a certain volume. For additional volume, you charge a usage fee that is aligned to the business value that your webhooks provide. At high volumes, you offer a premium tier where you provide dedicated infrastructure for certain consumers.

Monitoring and troubleshooting: Beyond the architecture, consider processes for day-to-day operations. As endpoints are managed by external parties, consider enabling self-service. For example, allow consumers to view statuses, replay events, and search for past webhook logs to diagnose issues.

Advanced Scenarios: This example is designed for popular use cases. For advanced scenarios, consider alternative application integration services noting their Service Quotas. For example, Amazon Simple Notification Service (SNS) for fan-out to a larger number of consumers, Lambda for flexibility to customize payloads and authentication, and AWS Step Functions for orchestrating a circuit breaker pattern to deactivate unreliable subscribers.

Receiving webhooks

To receive webhooks, you require an API to provide to the webhook provider. For example, an ecommerce store (consumer) may rely on notifications provided by their payment platform (provider) to ensure that goods are shipped in a timely manner. Webhooks present a unique scenario as the consumer must be scalable, resilient, and ensure that all requests are received.

AWS reference architecture for a webhook consumer

In this scenario, consider an advanced use case that can handle large payloads by using the claim-check pattern.

At a high-level, the architecture consists of:

API: An API endpoint to receive webhooks. An event-driven system then authorizes and processes the received webhooks.
Payload Store: S3 provides scalable storage for large payloads.
Webhook Processing: EventBridge Pipes provide an extensible architecture for processing. It can batch, filter, enrich, and send events to a range of processing services as targets.

Considerations and best practices for receiving webhooks

When building an application to receive webhooks, consider the following factors:

Scalability: Providers typically send events as they occur. API Gateway provides a scalable managed endpoint to receive events. If unavailable or throttled, providers may retry the request, however, this is not guaranteed. Therefore, it is important to configure appropriate rate and burst limits. Throttling requests at the entry point mitigates impact on downstream services, where each service has its own quotas and limits. In many cases, providers are also aware of impact on downstream systems. As such, they send events at a threshold rate limit, typically up to 500 transactions per second (TPS).

In addition, API Gateway allows you to validate requests, monitor for any errors, and protect against distributed denial of service (DDoS). This includes Layer 7 and Layer 3 attacks, which are common threats to webhook consumers given public exposure.

Authorization and Verification: Providers can support different authorization methods. Consider a common scenario with Hash-based Message Authentication Code (HMAC), where a shared secret is established and stored in Secrets Manager. A Lambda function then verifies integrity of the message, processing a signature in the request header. Typically, the signature contains a timestamped nonce with an expiry to mitigate replay attacks, where events are sent multiple times by an attacker. Alternatively, if the provider supports OAuth, consider securing the API with Amazon Cognito.

Payload Size: Providers may send a variety of payload sizes. Events can be batched to a single larger request, or they may contain significant information. Consider payload size limits in your event-driven system. API Gateway and Lambda have limits of 10 Mb and 6 Mb. However, DynamoDB and SQS are limited to 400kb and 256kb (with extension for large messages) which can represent a bottleneck.

Instead of processing the entire payload, S3 stores the payload. It is then referenced in DynamoDB, via its bucket name and object key. This is known as the claim-check pattern. With this approach, the architecture supports payloads of up to 6mb, as per the Lambda invocation payload quota.

Idempotency: For reliability, many providers prioritize delivering at-least-once, even if it means not guaranteeing exactly once delivery. They can transmit the same request multiple times, resulting in duplicates. To handle this, a Lambda function checks against the event’s unique identifier against previous records in DynamoDB. If not already processed, you create a DynamoDB item.

Ordering: Consider processing requests in its intended order. As most providers prioritize at-least-once delivery, events can be out of order. To indicate order, events may include a timestamp or a sequence identifier in the payload. If not, ordering may be on a best-efforts basis based on when the webhook is received. To handle ordering reliably, select event-driven services that ensure ordering. This example uses DynamoDB Streams and EventBridge Pipes.

Flexible Processing: EventBridge Pipes provide integrations to a range of event-driven services as targets. You can route events to different targets based on filters. Different event types may require different processors. For example, you can use Step Functions for orchestrating complex workflows, Lambda for compute operations with less than 15-minute execution time, SQS to buffer requests, and Amazon Elastic Container Service (ECS) for long-running compute jobs. EventBridge Pipes provide transformation to ensure only necessary payloads are sent, and enrichment if additional information is required.

Costs: This example considers a use case that can handle large payloads. However, if you can ensure that providers send minimal payloads, consider a simpler architecture without the claim-check pattern to minimize cost.

Conclusion

Webhooks are a popular method for applications to communicate, and for businesses to collaborate and integrate with customers and partners.

This post shows how you can build applications to send and receive webhooks on AWS. It uses serverless services such as EventBridge and Lambda, which are well-suited for event-driven use cases. It covers high-level reference architectures, considerations, best practices and code sample to assist in building your solution.

For standards and best practices on webhooks, visit the open-source community resources Webhooks.fyi and CloudEvents.io.

For more serverless learning resources, visit Serverless Land.

Unlock scalable analytics with AWS Glue and Google BigQuery

2023-10-28 Kartikay Khator

Post Syndicated from Kartikay Khator original https://aws.amazon.com/blogs/big-data/unlock-scalable-analytics-with-aws-glue-and-google-bigquery/

Data integration is the foundation of robust data analytics. It encompasses the discovery, preparation, and composition of data from diverse sources. In the modern data landscape, accessing, integrating, and transforming data from diverse sources is a vital process for data-driven decision-making. AWS Glue, a serverless data integration and extract, transform, and load (ETL) service, has revolutionized this process, making it more accessible and efficient. AWS Glue eliminates complexities and costs, allowing organizations to perform data integration tasks in minutes, boosting efficiency.

This blog post explores the newly announced managed connector for Google BigQuery and demonstrates how to build a modern ETL pipeline with AWS Glue Studio without writing code.

Overview of AWS Glue

AWS Glue is a serverless data integration service that makes it easier to discover, prepare, and combine data for analytics, machine learning (ML), and application development. AWS Glue provides all the capabilities needed for data integration, so you can start analyzing your data and putting it to use in minutes instead of months. AWS Glue provides both visual and code-based interfaces to make data integration easier. Users can more easily find and access data using the AWS Glue Data Catalog. Data engineers and ETL (extract, transform, and load) developers can visually create, run, and monitor ETL workflows in a few steps in AWS Glue Studio. Data analysts and data scientists can use AWS Glue DataBrew to visually enrich, clean, and normalize data without writing code.

Introducing Google BigQuery Spark connector

To meet the demands of diverse data integration use cases, AWS Glue now offers a native spark connector for Google BigQuery. Customers can now use AWS Glue 4.0 for Spark to read from and write to tables in Google BigQuery. Additionally, you can read an entire table or run a custom query and write your data using direct and indirect writing methods. You connect to BigQuery using service account credentials stored securely in AWS Secrets Manager.

Benefits of Google BigQuery Spark connector

Seamless integration: The native connector offers an intuitive and streamlined interface for data integration, reducing the learning curve.
Cost efficiency: Building and maintaining custom connectors can be expensive. The native connector provided by AWS Glue is a cost-effective alternative.
Efficiency: Data transformation tasks that previously took weeks or months can now be accomplished within minutes, optimizing efficiency.

Solution overview

In this example, you create two ETL jobs using AWS Glue with the native Google BigQuery connector.

Query a BigQuery table and save the data into Amazon Simple Storage Service (Amazon S3) in Parquet format.
Use the data extracted from the first job to transform and create an aggregated result to be stored in Google BigQuery.

Prerequisites

The dataset used in this solution is the NCEI/WDS Global Significant Earthquake Database, with a global listing of over 5,700 earthquakes from 2150 BC to the present. Copy this public data into your Google BigQuery project or use your existing dataset.

Configure BigQuery connections

To connect to Google BigQuery from AWS Glue, see Configuring BigQuery connections. You must create and store your Google Cloud Platform credentials in a Secrets Manager secret, then associate that secret with a Google BigQuery AWS Glue connection.

Set up Amazon S3

Every object in Amazon S3 is stored in a bucket. Before you can store data in Amazon S3, you must create an S3 bucket to store the results.

To create an S3 bucket:

On the AWS Management Console for Amazon S3, choose Create bucket.
Enter a globally unique Name for your bucket; for example, awsglue-demo.
Choose Create bucket.

Create an IAM role for the AWS Glue ETL job

When you create the AWS Glue ETL job, you specify an AWS Identity and Access Management (IAM) role for the job to use. The role must grant access to all resources used by the job, including Amazon S3 (for any sources, targets, scripts, driver files, and temporary directories), and Secrets Manager.

For instructions, see Configure an IAM role for your ETL job.

Solution walkthrough

Create a visual ETL job in AWS Glue Studio to transfer data from Google BigQuery to Amazon S3

Open the AWS Glue console.
In AWS Glue, navigate to Visual ETL under the ETL jobs section and create a new ETL job using Visual with a blank canvas.
Enter a Name for your AWS Glue job, for example, bq-s3-dataflow.
Select Google BigQuery as the data source.
1. Enter a name for your Google BigQuery source node, for example, noaa_significant_earthquakes.
2. Select a Google BigQuery connection, for example, bq-connection.
3. Enter a Parent project, for example, bigquery-public-datasources.
4. Select Choose a single table for the BigQuery Source.
5. Enter the table you want to migrate in the form [dataset].[table], for example, noaa_significant_earthquakes.earthquakes.
Next, choose the data target as Amazon S3.
1. Enter a Name for the target Amazon S3 node, for example, earthquakes.
2. Select the output data Format as Parquet.
3. Select the Compression Type as Snappy.
4. For the S3 Target Location, enter the bucket created in the prerequisites, for example, s3://<YourBucketName>/noaa_significant_earthquakes/earthquakes/.
5. You should replace <YourBucketName> with the name of your bucket.
Next go to the Job details. In the IAM Role, select the IAM role from the prerequisites, for example, AWSGlueRole.
Choose Save.

Run and monitor the job

After your ETL job is configured, you can run the job. AWS Glue will run the ETL process, extracting data from Google BigQuery and loading it into your specified S3 location.
Monitor the job’s progress in the AWS Glue console. You can see logs and job run history to ensure everything is running smoothly.

Data validation

After the job has run successfully, validate the data in your S3 bucket to ensure it matches your expectations. You can see the results using Amazon S3 Select.

Automate and schedule

If needed, set up job scheduling to run the ETL process regularly. You can use AWS to automate your ETL jobs, ensuring your S3 bucket is always up to date with the latest data from Google BigQuery.

You’ve successfully configured an AWS Glue ETL job to transfer data from Google BigQuery to Amazon S3. Next, you create the ETL job to aggregate this data and transfer it to Google BigQuery.

Finding earthquake hotspots with AWS Glue Studio Visual ETL.

Open AWS Glue console.
In AWS Glue navigate to Visual ETL under the ETL jobs section and create a new ETL job using Visual with a blank canvas.
Provide a name for your AWS Glue job, for example, s3-bq-dataflow.
Choose Amazon S3 as the data source.
1. Enter a Name for the source Amazon S3 node, for example, earthquakes.
2. Select S3 location as the S3 source type.
3. Enter the S3 bucket created in the prerequisites as the S3 URL, for example, s3://<YourBucketName>/noaa_significant_earthquakes/earthquakes/.
4. You should replace <YourBucketName> with the name of your bucket.
5. Select the Data format as Parquet.
6. Select Infer schema.
Next, choose Select Fields transformation.
1. Select earthquakes as Node parents.
2. Select fields: id, eq_primary, and country.
Next, choose Aggregate transformation.
1. Enter a Name, for example Aggregate.
2. Choose Select Fields as Node parents.
3. Choose eq_primary and country as the group by columns.
4. Add id as the aggregate column and count as the aggregation function.
Next, choose RenameField transformation.
1. Enter a name for the source Amazon S3 node, for example, Rename eq_primary.
2. Choose Aggregate as Node parents.
3. Choose eq_primary as the Current field name and enter earthquake_magnitude as the New field name.
Next, choose RenameField transformation
1. Enter a name for the source Amazon S3 node, for example, Rename count(id).
2. Choose Rename eq_primary as Node parents.
3. Choose count(id) as the Current field name and enter number_of_earthquakes as the New field name.
Next, choose the data target as Google BigQuery.
1. Provide a name for your Google BigQuery source node, for example, most_powerful_earthquakes.
2. Select a Google BigQuery connection, for example, bq-connection.
3. Select Parent project, for example, bigquery-public-datasources.
4. Enter the name of the Table you want to create in the form [dataset].[table], for example, noaa_significant_earthquakes.most_powerful_earthquakes.
5. Choose Direct as the Write method.
Next go to the Job details tab and in the IAM Role, select the IAM role from the prerequisites, for example, AWSGlueRole.
Choose Save.

Run and monitor the job

After your ETL job is configured, you can run the job. AWS Glue runs the ETL process, extracting data from Google BigQuery and loading it into your specified S3 location.
Monitor the job’s progress in the AWS Glue console. You can see logs and job run history to ensure everything is running smoothly.

Data validation

After the job has run successfully, validate the data in your Google BigQuery dataset. This ETL job returns a list of countries where the most powerful earthquakes have occurred. It provides these by counting the number of earthquakes for a given magnitude by country.

Automate and schedule

You can set up job scheduling to run the ETL process regularly. AWS Glue allows you to automate your ETL jobs, ensuring your S3 bucket is always up to date with the latest data from Google BigQuery.

That’s it! You’ve successfully set up an AWS Glue ETL job to transfer data from Amazon S3 to Google BigQuery. You can use this integration to automate the process of data extraction, transformation, and loading between these two platforms, making your data readily available for analysis and other applications.

Clean up

To avoid incurring charges, clean up the resources used in this blog post from your AWS account by completing the following steps:

On the AWS Glue console, choose Visual ETL in the navigation pane.
From the list of jobs, select the job bq-s3-data-flow and delete it.
From the list of jobs, select the job s3-bq-data-flow and delete it.
On the AWS Glue console, choose Connections in the navigation pane under Data Catalog.
Choose the BiqQuery connection you created and delete it.
On the Secrets Manager console, choose the secret you created and delete it.
On the IAM console, choose Roles in the navigation pane, then select the role you created for the AWS Glue ETL job and delete it.
On the Amazon S3 console, search for the S3 bucket you created, choose Empty to delete the objects, then delete the bucket.
Clean up resources in your Google account by deleting the project that contains the Google BigQuery resources. Follow the documentation to clean up the Google resources.

Conclusion

The integration of AWS Glue with Google BigQuery simplifies the analytics pipeline, reduces time-to-insight, and facilitates data-driven decision-making. It empowers organizations to streamline data integration and analytics. The serverless nature of AWS Glue means no infrastructure management, and you pay only for the resources consumed while your jobs are running. As organizations increasingly rely on data for decision-making, this native spark connector provides an efficient, cost-effective, and agile solution to swiftly meet data analytics needs.

If you’re interested to see how to read from and write to tables in Google BigQuery in AWS Glue, take a look at step-by-step video tutorial. In this tutorial, we walk through the entire process, from setting up the connection to running the data transfer flow. For more information on AWS Glue, visit AWS Glue.

Appendix

If you are looking to implement this example, using code instead of the AWS Glue console, use the following code snippets.

Reading data from Google BigQuery and writing data into Amazon S3

import sys
from awsglue.transforms import *
from awsglue.utils import getResolvedOptions
from pyspark.context import SparkContext
from awsglue.context import GlueContext
from awsglue.job import Job

args = getResolvedOptions(sys.argv, ["JOB_NAME"])
sc = SparkContext()
glueContext = GlueContext(sc)
spark = glueContext.spark_session
job = Job(glueContext)
job.init(args["JOB_NAME"], args)

# STEP-1 Read the data from Big Query Table 
noaa_significant_earthquakes_node1697123333266 = (
    glueContext.create_dynamic_frame.from_options(
        connection_type="bigquery",
        connection_options={
            "connectionName": "bq-connection",
            "parentProject": "bigquery-public-datasources",
            "sourceType": "table",
            "table": "noaa_significant_earthquakes.earthquakes",
        },
        transformation_ctx="noaa_significant_earthquakes_node1697123333266",
    )
)
# STEP-2 Write the data read from Big Query Table into S3
# You should replace <YourBucketName> with the name of your bucket.
earthquakes_node1697157772747 = glueContext.write_dynamic_frame.from_options(
    frame=noaa_significant_earthquakes_node1697123333266,
    connection_type="s3",
    format="glueparquet",
    connection_options={
        "path": "s3://<YourBucketName>/noaa_significant_earthquakes/earthquakes/",
        "partitionKeys": [],
    },
    format_options={"compression": "snappy"},
    transformation_ctx="earthquakes_node1697157772747",
)

job.commit()

Reading and aggregating data from Amazon S3 and writing into Google BigQuery

import sys
from awsglue.transforms import *
from awsglue.utils import getResolvedOptions
from pyspark.context import SparkContext
from awsglue.context import GlueContext
from awsglue.job import Job
from awsglue.dynamicframe import DynamicFrame
from awsglue import DynamicFrame
from pyspark.sql import functions as SqlFuncs

def sparkAggregate(
    glueContext, parentFrame, groups, aggs, transformation_ctx
) -> DynamicFrame:
    aggsFuncs = []
    for column, func in aggs:
        aggsFuncs.append(getattr(SqlFuncs, func)(column))
    result = (
        parentFrame.toDF().groupBy(*groups).agg(*aggsFuncs)
        if len(groups) > 0
        else parentFrame.toDF().agg(*aggsFuncs)
    )
    return DynamicFrame.fromDF(result, glueContext, transformation_ctx)

args = getResolvedOptions(sys.argv, ["JOB_NAME"])
sc = SparkContext()
glueContext = GlueContext(sc)
spark = glueContext.spark_session
job = Job(glueContext)
job.init(args["JOB_NAME"], args)

# STEP-1 Read the data from Amazon S3 bucket
# You should replace <YourBucketName> with the name of your bucket.
earthquakes_node1697218776818 = glueContext.create_dynamic_frame.from_options(
    format_options={},
    connection_type="s3",
    format="parquet",
    connection_options={
        "paths": [
            "s3://<YourBucketName>/noaa_significant_earthquakes/earthquakes/"
        ],
        "recurse": True,
    },
    transformation_ctx="earthquakes_node1697218776818",
)

# STEP-2 Select fields
SelectFields_node1697218800361 = SelectFields.apply(
    frame=earthquakes_node1697218776818,
    paths=["id", "eq_primary", "country"],
    transformation_ctx="SelectFields_node1697218800361",
)

# STEP-3 Aggregate data
Aggregate_node1697218823404 = sparkAggregate(
    glueContext,
    parentFrame=SelectFields_node1697218800361,
    groups=["eq_primary", "country"],
    aggs=[["id", "count"]],
    transformation_ctx="Aggregate_node1697218823404",
)

Renameeq_primary_node1697219483114 = RenameField.apply(
    frame=Aggregate_node1697218823404,
    old_name="eq_primary",
    new_name="earthquake_magnitude",
    transformation_ctx="Renameeq_primary_node1697219483114",
)

Renamecountid_node1697220511786 = RenameField.apply(
    frame=Renameeq_primary_node1697219483114,
    old_name="`count(id)`",
    new_name="number_of_earthquakes",
    transformation_ctx="Renamecountid_node1697220511786",
)

# STEP-1 Write the aggregated data in Google BigQuery
most_powerful_earthquakes_node1697220563923 = (
    glueContext.write_dynamic_frame.from_options(
        frame=Renamecountid_node1697220511786,
        connection_type="bigquery",
        connection_options={
            "connectionName": "bq-connection",
            "parentProject": "bigquery-public-datasources",
            "writeMethod": "direct",
            "table": "noaa_significant_earthquakes.most_powerful_earthquakes",
        },
        transformation_ctx="most_powerful_earthquakes_node1697220563923",
    )
)

job.commit()

About the authors

Kartikay Khator is a Solutions Architect in Global Life Sciences at Amazon Web Services (AWS). He is passionate about building innovative and scalable solutions to meet the needs of customers, focusing on AWS Analytics services. Beyond the tech world, he is an avid runner and enjoys hiking.

Kamen Sharlandjiev is a Sr. Big Data and ETL Solutions Architect and Amazon AppFlow expert. He’s on a mission to make life easier for customers who are facing complex data integration challenges. His secret weapon? Fully managed, low-code AWS services that can get the job done with minimal effort and no coding.

Anshul Sharma is a Software Development Engineer in AWS Glue Team. He is driving the connectivity charter which provide Glue customer native way of connecting any Data source (Data-warehouse, Data-lakes, NoSQL etc) to Glue ETL Jobs. Beyond the tech world, he is a cricket and soccer lover.