Tag Archives: Application Integration

Enriching Event-Driven Architectures with AWS Event Fork Pipelines

Post Syndicated from Rachel Richardson original https://aws.amazon.com/blogs/compute/enriching-event-driven-architectures-with-aws-event-fork-pipelines/

This post is courtesy of Otavio Ferreira, Mgr, Amazon SNS, and James Hood, Sr. Software Dev Engineer

Many customers are choosing to build event-driven applications in which subscriber services automatically perform work in response to events triggered by publisher services. This architectural pattern can make services more reusable, interoperable, and scalable.

Customers often fork event processing into pipelines that address common event handling requirements, such as event storage, backup, search, analytics, or replay. To help you build event-driven applications even faster, AWS introduces Event Fork Pipelines, a collection of open-source event handling pipelines that you can subscribe to Amazon SNS topics in your AWS account.

Event Fork Pipelines is a suite of open-source nested applications, based on the AWS Serverless Application Model (AWS SAM). You can deploy it directly from the AWS Serverless Application Repository into your AWS account.

Event Fork Pipelines is built on top of serverless services, including Amazon SNS, Amazon SQS, and AWS Lambda. These services provide serverless building blocks that help you build fully managed, highly available, and scalable event-driven platforms. Lambda enables you to build event-driven microservices as serverless functions. SNS and SQS provide serverless topics and queues for integrating these microservices and other distributed systems in your architecture. These building blocks are at the core of the modern application development best practices.

Surfacing the event fork pattern

At AWS, we’ve worked closely with customers across market segments and geographies on event-driven architectures. For example:

  • Financial platforms that handle events related to bank transactions and stock ticks
  • Retail platforms that trigger checkout and fulfillment events

At scale, event-driven architectures often require a set of supporting services to address common requirements such as system auditability, data discoverability, compliance, business insights, and disaster recovery. Translated to AWS, customers often connect event-driven applications to services such as Amazon S3 for event storage and backup, and to Amazon Elasticsearch Service for event search and analytics. Also, customers often implement an event replay mechanism to recover from failure modes in their applications.

AWS created Event Fork Pipelines to encapsulate these common requirements, reducing the amount of effort required for you to connect your event-driven architectures to these supporting AWS services.

AWS then started sharing this pattern more broadly, so more customers could benefit. At the 2018 AWS re:Invent conference in Las Vegas, Amazon CTO Werner Vogels announced the launch of nested applications in his keynote. Werner shared the Event Fork Pipelines pattern with the audience as an example of common application logic that had been encapsulated as a set of nested applications.

The following reference architecture diagram shows an application supplemented by three nested applications:

Each pipeline is subscribed to the same SNS topic, and can process events in parallel as these events are published to the topic. Each pipeline is independent and can set its own subscription filter policy. That way, it processes only the subset of events that it’s interested in, rather than all events published to the topic.

Amazon SNS Fork pipelines reference architecture

Figure 1 – Reference architecture using Event Fork Pipelines

The three event fork pipelines are placed alongside your regular event processing pipelines, which are potentially already subscribed to your SNS topic. Therefore, you don’t have to change any portion of your current message publisher to take advantage of Event Fork Pipelines in your existing workloads. The following sections describe these pipelines and how to deploy them in your system architecture.

Understanding the catalog of event fork pipelines

In the abstract, Event Fork Pipelines is a serverless design pattern. Concretely, Event Fork Pipelines is also a suite of nested serverless applications, based on AWS SAM. You deploy the nested applications directly from the AWS Serverless Application Repository to your AWS account, to enrich your event-driven platforms. You can deploy them individually in your architecture, as needed.

Here’s more information about each nested application in the Event Fork Pipelines suite.

Event Storage & Backup pipeline

Event Fork Pipeline for Event Storage & Backup

Figure 2 – Event Fork Pipeline for Event Storage & Backup

The preceding diagram shows the Event Storage & Backup pipeline. You can subscribe this pipeline to your SNS topic to automatically back up the events flowing through your system. This pipeline is composed of the following resources:

  • An SQS queue that buffers the events delivered by the SNS topic
  • A Lambda function that automatically polls for these events in the queue and pushes them into an Amazon Kinesis Data Firehose delivery stream
  • An S3 bucket that durably backs up the events loaded by the stream

You can configure this pipeline to fine-tune the behavior of your delivery stream. For example, you can configure your pipeline so that the underlying delivery stream buffers, transforms, and compresses your events before loading them into the bucket. As events are loaded, you can use Amazon Athena to query the bucket using standard SQL queries. Also, you can configure the pipeline to either reuse an existing S3 bucket or create a new one for you.

Event Search & Analytics pipeline

Event Fork Pipeline for Event Search & Analytics

Figure 3 – Event Fork Pipeline for Event Search & Analytics

The preceding diagram shows the Event Search & Analytics pipeline. You can subscribe this pipeline to your SNS topic to index in a search domain the events flowing through your system, and then run analytics on them. This pipeline is composed of the following resources:

  • An SQS queue that buffers the events delivered by the SNS topic
  • A Lambda function that polls events from the queue and pushes them into a Data Firehose delivery stream
  • An Amazon ES domain that indexes the events loaded by the delivery stream
  • An S3 bucket that stores the dead-letter events that couldn’t be indexed in the search domain

You can configure this pipeline to fine-tune your delivery stream in terms of event buffering, transformation and compression. You can also decide whether the pipeline should reuse an existing Amazon ES domain in your AWS account or create a new one for you. As events are indexed in the search domain, you can use Kibana to run analytics on your events and update visual dashboards in real time.

Event Replay pipeline

Event Fork Pipeline for Event Replay

Figure 4 – Event Fork Pipeline for Event Replay

The preceding diagram shows the Event Replay pipeline. You can subscribe this pipeline to your SNS topic to record the events that have been processed by your system for up to 14 days. You can then reprocess them in case your platform is recovering from a failure or a disaster. This pipeline is composed of the following resources:

  • An SQS queue that buffers the events delivered by the SNS topic
  • A Lambda function that polls events from the queue and redrives them into your regular event processing pipeline, which is also subscribed to your topic

By default, the replay function is disabled, which means it isn’t redriving your events. If the events need to be reprocessed, your operators must enable the replay function.

Applying event fork pipelines in a use case

This is how everything comes together. The following scenario describes an event-driven, serverless ecommerce application that uses the Event Fork Pipelines pattern. This example ecommerce application is available in AWS Serverless Application Repository. You can deploy it to your AWS account using the Lambda console, test it, and look at its source code in GitHub.

Example ecommerce application using Event Fork Pipelines

Figure 5 – Example e-commerce application using Event Fork Pipelines

The ecommerce application takes orders from buyers through a RESTful API hosted by Amazon API Gateway and backed by a Lambda function named CheckoutFunction. This function publishes all orders received to an SNS topic named CheckoutEventsTopic, which in turn fans out the orders to four different pipelines. The first pipeline is the regular checkout-processing pipeline designed and implemented by you as the ecommerce application owner. This pipeline has the following resources:

  • An SQS queue named CheckoutQueue that buffers all orders received
  • A Lambda function named CheckoutFunction that polls the queue to process these orders
  • An Amazon DynamoDB table named CheckoutTable that securely saves all orders as they’re placed

The components of the system described thus far handle what you might think of as the core business logic. But in addition, you should address the set of elements necessary for making the system resilient, compliant, and searchable:

  • Backing up all orders securely. Compressed backups must be encrypted at rest, with sensitive payment details removed for security and compliance purposes.
  • Searching and running analytics on orders, if the amount is $100 or more. Analytics are needed for key ecommerce metrics, such as average ticket size, average shipping time, most popular products, and preferred payment options.
  • Replaying recent orders. If the fulfillment process is disrupted at any point, you should be able to replay the most recent orders from up to two weeks. This is a key requirement that guarantees the continuity of the ecommerce business.

Rather than implementing all the event processing logic yourself, you can choose to subscribe Event Fork Pipelines to your existing SNS topic CheckoutEventsTopic. The pipelines are configured as follows:

  • The Event Storage & Backup pipeline is configured to transform data as follows:
    • Remove credit card details
    • Buffer data for 60 seconds
    • Compress data using GZIP
    • Encrypt data using the default customer master key (CMK) for S3

This CMK is managed by AWS and powered by AWS Key Management Service (AWS KMS). For more information, see Choosing Amazon S3 for Your Destination, Data Transformation, and Configuration Settings in the Amazon Kinesis Data Firehose Developer Guide.

  • The Event Search & Analytics pipeline is configured with:
    • An index retry duration of 30 seconds
    • A bucket for storing orders that failed to be indexed in the search domain
    • A filter policy to restrict the set of orders that are indexed

For more information, see Choosing Amazon ES for Your Destination, in the Amazon Kinesis Data Firehose Developer Guide.

  • The Event Replay pipeline is configured with the SQS queue name that is part of the regular checkout processing pipeline. For more information, see Queue Name and URL in the Amazon SQS Developer Guide.

The filter policy, shown in JSON format, is set in the configuration for the Event Search & Analytics pipeline. This filter policy matches only incoming orders in which the total amount is $100 or more. For more information, see Message Filtering in the Amazon SNS Developer Guide.


    "amount": [

        { "numeric": [ ">=", 100 ] }



By using the Event Fork Pipelines pattern, you avoid the development overhead associated with coding undifferentiated logic for handling events.

Event Fork Pipelines can be deployed directly from AWS Serverless Application Repository into your AWS account.

Deploying event fork pipelines

Event Fork Pipelines is available as a set of public apps in the AWS Serverless Application Repository (to find the apps, select the ‘Show apps that create custom IAM roles or resource policies’ check box under the search bar). It can be deployed and tested manually via the Lambda console. In a production scenario, we recommend embedding fork pipelines within the AWS SAM template of your overall application. The nested applications feature enables you to do this by adding an AWS::Serverless::Application resource to your AWS SAM template. The resource references the ApplicationId and SemanticVersion values of the application to nest.

For example, you can include the Event Storage & Backup pipeline as a nested application by adding the following YAML snippet to the Resources section of your AWS SAM template:


  Type: AWS::Serverless::Application



      ApplicationId: arn:aws:serverlessrepo:us-east-1:012345678901:applications/fork-event-storage-backup-pipeline

      SemanticVersion: 1.0.0


      # SNS topic ARN whose messages should be backed up to the S3 bucket.

      TopicArn: !Ref MySNSTopic

When specifying parameter values, you can use AWS CloudFormation intrinsic functions to reference other resources in your template. In the preceding example, the TopicArn parameter is filled in by referencing an AWS::SNS::Topic called MySNSTopic, defined elsewhere in the AWS SAM template. For more information, see Intrinsic Function Reference in the AWS CloudFormation User Guide.

To copy the YAML required for nesting, in the Lambda console page for an AWS Serverless Application Repository application, choose Copy as SAM Resource.

Authoring new event fork pipelines

We invite you to fork the Event Fork Pipelines repository in GitHub and submit pull requests for contributing with new pipelines. In addition to event storage and backup, event search and analytics, and event replay, what other common event handling requirements have you seen?

We look forward to seeing what you’ll come up with for extending the Event Fork Pipelines suite.


Event Fork Pipelines is a serverless design pattern and a suite of open-source nested serverless applications, based on AWS SAM. You can deploy it directly from AWS Serverless Application Repository to enrich your event-driven system architecture. Event Fork Pipelines lets you store, back up, replay, search, and run analytics on the events flowing through your system. There’s no need to write code, manually stitch resources together, or set up infrastructure.

You can deploy Event Fork Pipelines in any AWS Region that supports the underlying AWS services used in the pipelines. There are no additional costs associated with Event Fork Pipelines itself, and you pay only for using the AWS resources inside each nested application.

Get started today by deploying the example ecommerce application or searching for Event Fork Pipelines in AWS Serverless Application Repository.

Learn about AWS Services & Solutions – February 2019 AWS Online Tech Talks

Post Syndicated from Robin Park original https://aws.amazon.com/blogs/aws/learn-about-aws-services-solutions-february-2019-aws-online-tech-talks/

AWS Tech Talks

Join us this February to learn about AWS services and solutions. The AWS Online Tech Talks are live, online presentations that cover a broad range of topics at varying technical levels. These tech talks, led by AWS solutions architects and engineers, feature technical deep dives, live demonstrations, customer examples, and Q&A with AWS experts. Register Now!

Note – All sessions are free and in Pacific Time.

Tech talks this month:

Application Integration

February 20, 2019 | 11:00 AM – 12:00 PM PTCustomer Showcase: Migration & Messaging for Mission Critical Apps with S&P Global Ratings – Learn how S&P Global Ratings meets the high availability and fault tolerance requirements of their mission critical applications using the Amazon MQ.


February 28, 2019 | 1:00 PM – 2:00 PM PTBuild AR/VR Apps with AWS: Creating a Multiplayer Game with Amazon Sumerian – Learn how to build real-world augmented reality, virtual reality and 3D applications with Amazon Sumerian.


February 18, 2019 | 11:00 AM – 12:00 PM PTDeep Dive on Amazon Managed Blockchain – Explore the components of blockchain technology, discuss use cases, and do a deep dive into capabilities, performance, and key innovations in Amazon Managed Blockchain.


February 25, 2019 | 9:00 AM – 10:00 AM PTWhat’s New in Amazon EC2 – Learn about the latest innovations in Amazon EC2, including new instances types, related technologies, and consumption options that help you optimize running your workloads for performance and cost.

February 27, 2019 | 1:00 PM – 2:00 PM PTDeploy and Scale Your First Cloud Application with Amazon Lightsail – Learn how to quickly deploy and scale your first multi-tier cloud application using Amazon Lightsail.


February 19, 2019 | 9:00 AM – 10:00 AM PTSecuring Container Workloads on AWS Fargate – Explore the security controls and best practices for securing containers running on AWS Fargate.

Data Lakes & Analytics

February 18, 2019 | 1:00 PM – 2:00 PM PTAmazon Redshift Tips & Tricks: Scaling Storage and Compute Resources – Learn about the tools and best practices Amazon Redshift customers can use to scale storage and compute resources on-demand and automatically to handle growing data volume and analytical demand.


February 18, 2019 | 9:00 AM – 10:00 AM PTBuilding Real-Time Applications with Redis – Learn about Amazon’s fully managed Redis service and how it makes it easier, simpler, and faster to build real-time applications.

February 21, 2019 | 1:00 PM – 2:00 PM PT – Introduction to Amazon DocumentDB (with MongoDB Compatibility) – Get an introduction to Amazon DocumentDB (with MongoDB compatibility), a fast, scalable, and highly available document database that makes it easy to run, manage & scale MongoDB-workloads.


February 20, 2019 | 1:00 PM – 2:00 PM PTFireside Chat: DevOps at Amazon with Ken Exner, GM of AWS Developer Tools – Join our fireside chat with Ken Exner, GM of Developer Tools, to learn about Amazon’s DevOps transformation journey and latest practices and tools that support the current DevOps model.

End-User Computing

February 28, 2019 | 9:00 AM – 10:00 AM PTEnable Your Remote and Mobile Workforce with Amazon WorkLink – Learn about Amazon WorkLink, a new, fully-managed service that provides your employees secure, one-click access to internal corporate websites and web apps using their mobile phones.

Enterprise & Hybrid

February 26, 2019 | 1:00 PM – 2:00 PM PTThe Amazon S3 Storage Classes – For cloud ops professionals, by cloud ops professionals. Wallace and Orion will tackle your toughest AWS hybrid cloud operations questions in this live Office Hours tech talk.


February 26, 2019 | 9:00 AM – 10:00 AM PTBring IoT and AI Together – Learn how to bring intelligence to your devices with the intersection of IoT and AI.

Machine Learning

February 19, 2019 | 1:00 PM – 2:00 PM PTGetting Started with AWS DeepRacer – Learn about the basics of reinforcement learning, what’s under the hood and opportunities to get hands on with AWS DeepRacer and how to participate in the AWS DeepRacer League.

February 20, 2019 | 9:00 AM – 10:00 AM PTBuild and Train Reinforcement Models with Amazon SageMaker RL – Learn about Amazon SageMaker RL to use reinforcement learning and build intelligent applications for your businesses.

February 21, 2019 | 11:00 AM – 12:00 PM PTTrain ML Models Once, Run Anywhere in the Cloud & at the Edge with Amazon SageMaker Neo – Learn about Amazon SageMaker Neo where you can train ML models once and run them anywhere in the cloud and at the edge.

February 28, 2019 | 11:00 AM – 12:00 PM PTBuild your Machine Learning Datasets with Amazon SageMaker Ground Truth – Learn how customers are using Amazon SageMaker Ground Truth to build highly accurate training datasets for machine learning quickly and reduce data labeling costs by up to 70%.


February 27, 2019 | 11:00 AM – 12:00 PM PTMaximize the Benefits of Migrating to the Cloud – Learn how to group and rationalize applications and plan migration waves in order to realize the full set of benefits that cloud migration offers.


February 27, 2019 | 9:00 AM – 10:00 AM PTSimplifying DNS for Hybrid Cloud with Route 53 Resolver – Learn how to enable DNS resolution in hybrid cloud environments using Amazon Route 53 Resolver.

Productivity & Business Solutions

February 26, 2019 | 11:00 AM – 12:00 PM PTTransform the Modern Contact Center Using Machine Learning and Analytics – Learn how to integrate Amazon Connect and AWS machine learning services, such Amazon Lex, Amazon Transcribe, and Amazon Comprehend, to quickly process and analyze thousands of customer conversations and gain valuable insights.


February 19, 2019 | 11:00 AM – 12:00 PM PTBest Practices for Serverless Queue Processing – Learn the best practices of serverless queue processing, using Amazon SQS as an event source for AWS Lambda.


February 25, 2019 | 11:00 AM – 12:00 PM PT Introducing AWS Backup: Automate and Centralize Data Protection in the AWS Cloud – Learn about this new, fully managed backup service that makes it easy to centralize and automate the backup of data across AWS services in the cloud as well as on-premises.

New: Application integration with AWS Cloud Map for service discovery

Post Syndicated from AWS Admin original https://aws.amazon.com/blogs/architecture/new-application-integration-with-aws-cloud-map-for-service-discovery/

By: Alexandr Moroz, Sr. Product Manager, Amazon Route 53; Madhuri Peri, Sr. IoT Architect, AWS Professional Services; Aaron Molitor, Sr. Infrastructure Architect, AWS Professional Services; and Sarma Palli, Sr. DevOps Architect, AWS Professional Services

AWS Cloud Map enables you to map your cloud. You can define friendly names for any resource, such as Amazon S3 buckets, Amazon DynamoDB tables, Amazon SQS queues, or custom cloud services built on Amazon EC2, Amazon ECS, Amazon EKS, or AWS Lambda. Your applications can then discover resource location and metadata by friendly name using the AWS SDK and authenticated API queries. Resources can be further filtered and discovered by custom attributes such as deployment stage or version.

What’s new with API service discovery

If you want an enterprise application component such as a database hosted on Amazon EC2 instances to provide an endpoint to your database service, you have to register your applications’ EC2 IP address with AWS Cloud Map. You could register additional metadata attributes, like INSTANCE_STATUS, and then use this attribute with AWS Cloud Map to identify when the service is READY so that querying applications can only attempt a connection when they see a READY status in AWS Cloud Map. In cases where different microservices or enterprise applications have endpoints that have to be discovered, you can use AWS Cloud Map to register those as well. Examples of such endpoints include types of ELB load balancers, including ELB Classic, Application Load Balancers (ALB), and Network Load Balancers (NLB) with Auto Scaling groups.

Compute stack choices

Modern application architectures require a way to expose and advertise the service endpoint, register and de-register the endpoints, and query them. The dependencies of applications are expected to be handled by the applications themselves where a service registry becomes critical.

These microservices could follow different patterns of architecture lending themselves to use:

  1. Traditional workloads running on Amazon EC2 fronted by Auto Scaling groups or an ELB load balancer such as ELB Classic, Application Load Balancer, or Network Load Balancer.
  2. Amazon API Gateway and AWS Lambda for event-driven workflows.
  3. Container-based workloads on Amazon Elastic Container Service (ECS) using EC2 or Fargate launch types and Amazon Elastic Container Service for Kubernetes (EKS) for workloads that run as services (long-running) or daemons or run to completion (Batch / cron type).

This image shows a typical enterprise application composed of components that run different architectures. There is a web server running on Amazon EKS, a backend on Amazon ECS, a serverless event registration service, and payments running on EC2 Auto Scaling groups (ASG) while leveraging databases on Amazon Relational Database Service (RDS).


From a service discovery perspective, this is how the applications would want to be discovered and queried:

Let’s see how we can register each of these microservices (which are running on different cloud compute products) with AWS Cloud Map using both DNS-based and API-based service discovery and leveraging attributes for discovery when components are ready for traffic.

Microservice endpoints and discovery

AWS Cloud Map is a managed solution that lets you map logical names to the components/resources for an application. It allows applications to discover the resources using one of the AWS SDKs, RESTful API calls, or DNS queries. AWS Cloud Map serves registered resources, which can be Amazon DynamoDB tables, Amazon Simple Queue Service (SQS) queues, any higher-level application services that are built using EC2 instances or ECS tasks, or using a serverless stack.

When you register a resource, you can specify attributes and clients that can use the attributes to filter which resources are to be returned. For example, an application can request resources in a particular deployment stage, like Gamma or Prod. Additionally, you can choose to enable health checking for your IP-based resources, ensuring that AWS Cloud Map returns only healthy endpoints. Each API call is authenticated, and developers can control access to service locations and configuration using AWS Identity and Access Management (IAM).  This ensures that clients always discover the services that they’re authorized to use.

Let’s cover fundamentals

There are two aspects to service discovery:

  • The microservices themselves that register/de-register
  • Other microservices that are discover / query microservices

To register a microservice, follow these steps:

  1. Create a namespace.
  2. Create a service.
  3. Register instances with the service.

Steps 1 and 2  are performed once for each service. A utility function for registration and de-registration of a microservice has to be created. This utility function can be invoked for microservices regardless of the compute stack choice and deployed through your CI/CD/DevOps processes.

Step 3 is an ongoing operation that has to be updated each time the underlying EC2 compute that powers it changes. Examples include: EC2 Amazon Machine Image (AMI) changes, code changes for the service, and version changes.

Creating a namespace

A namespace is a logical group of services that share the same domain name, such as example1.example.com or example2.example.com. If you want these namespaces to be queried from only within your VPC, opt for a private namespace. If you want them to be accessible over the Internet, create a public namespace. In our example, the public namespace could be example1, but the total count of use of items in example1 in a tracker/reporting service could be an internal service.

Microservice using DNS-based service discovery:

Microservice using API-based service discovery:

Creating a service

When you register a service, AWS Cloud Map will create a record in the hosted zone – which is a combination of the name of the service and the name of the namespace. You could optionally define a health check for the service, too.

If the service you are creating is meant for DNS-based discovery using one of the A, AAAA, or SRV records, then you can create your service using the following syntax. Examples of this could be your application code running on an EC2 instance or as a container (ECS/EKS).

For services that are meant to be used only in an API-only namespace, the API call would look like this:

Register the compute backend with the service

Container IP address register/de-register

Amazon ECS is tightly integrated with AWS Cloud Map to enable service discovery for compute workloads running in ECS. When you enable service discovery for ECS services, it automatically keeps track of all task instances in AWS Cloud Map. Now your applications can discover them using DNS queries and AWS Cloud Map DiscoverInstances API calls. The ECS control plane that issues the calls would register the IP address of the task (and containers) with the AWS Cloud Map.

When the task goes away – either because a newer version has been deployed or there is a crash or a restart – the ECS control plane handles the de-registration process as well.

If you are using ECS for running containers, this is done seamlessly with ECS and AWS Cloud Map API integration.

API Gateway URL and AWS Lambda

When you create a microservice with an API namespace, you could use any attributes you prefer, without providing the IP/port information.

EC2 instance IP address registration and de-registration

As the EC2 instances are coming online, the userdata section of the bootstrap configuration will issue commands to register the EC2 instance’s IP address with the service. An alternate approach would be to run a Lambda function that runs against a microservice’s Auto Scaling group, lists the IP addresses, and registers the instance against the service.

If you are using an EC2 instance, if the instance is integrated with an Auto Scaling group, lifecycle hooks could also be used to run the de-register scripts. Another approach would be to use a Lambda function that runs periodically against an Auto Scaling group or even fires on Auto Scaling group notification events.


Both DNS and API service discovery are now supported by the AWS Cloud Map service discovery. Supported DNS record types are – A, AAAA, SRV, and CNAME.

It is typical in a microservices architecture for a service to be able to discover other services. We recommend that you query only by name and/or endpoint, and do not use the IP address of the compute stack (AWS Lambda / container/ EC2) that is backing the service.

The API commands list_services and get_services provide the information on what services are available and their corresponding details.

A DNS protocol also has clients caching the responses, so make sure that you handle caching settings. AWS Cloud Map uses regional endpoints here. Any A records created will use either a WEIGHTED response or MULTIVALUE answer policy. If you are using a Java-based compute stack, you might not want to choose DNS-based service discovery as the JVM caches DNS name lookups. When the JVM resolves a hostname to an IP address, it caches the IP address for a specified period of time, TTL. In such cases, you could use API-based service discovery and leverage the same approach as your other microservices that can use AWS Cloud Map.

DiscoverInstances API

DiscoverInstances API discovers registered instances for a specified namespace and service using regional endpoints. Updates to your services, such as new instances registered or existing instances removed, will be available in the API faster than via DNS. The API provides the ability to decorate the resources with additional metadata (service attributes) that can be used during discovery. For example, get the services with attributes of blue or green or other application attributes. These attributes can be used to complement health checks while performing discovery (such as finding out whether the instance is ready or not).

Here is a screenshot that shows how the registered ECS task instances appear in the AWS Cloud Map console:

The idea is that as the container or EC2 instance comes online or goes offline, it needs to issue a call to the AWS Cloud Map API to register or de-register the compute IP address.

Get started by visiting the AWS Cloud Map page. To learn more, take a look at the demo code in the GitHub repo here. If your compute workloads use EKS, please refer to this blog post that shows how to make EKS automatically publish all services in AWS Cloud Map.

Implementing enterprise integration patterns with AWS messaging services: point-to-point channels

Post Syndicated from Rachel Richardson original https://aws.amazon.com/blogs/compute/implementing-enterprise-integration-patterns-with-aws-messaging-services-point-to-point-channels/

This post is courtesy of Christian Mueller, Sr. Solutions Architect, AWS and Dirk Fröhner, Sr. Solutions Architect, AWS

At AWS, we see our customers increasingly moving toward managed services to reduce the time and money that they spend managing infrastructure. This also applies to the messaging domain, where AWS provides a collection of managed services.

Asynchronous messaging is a fundamental approach for integrating independent systems or building up a set of loosely coupled systems that can scale and evolve independently and flexibly. The well-known collection of enterprise integration patterns (EIPs) provides a “technology-independent vocabulary” to “design and document integration solutions.” This blog is the first of two that describes how you can implement the core EIPs using AWS messaging services. Let’s first look at the relevant AWS messaging services.

When organizations migrate their traditional messaging and existing applications to the cloud gradually, they usually want to do it without rewriting their code. Amazon MQ is a managed message broker service for Apache ActiveMQ that makes it easy to set up and operate message brokers in the cloud. It supports industry-standard APIs and protocols such as JMS, AMQP, and MQTT, so you can switch from any standards-based message broker to Amazon MQ without rewriting the messaging code in your applications. Amazon MQ is recommended if you’re using messaging with existing applications and want to move your messaging to the cloud without rewriting existing code.

However, if you build new applications for the cloud, we recommend that you consider using cloud-native messaging services such as Amazon SQS and Amazon SNS. These serverless, fully managed message queue and topic services scale to meet your demands and provide simple, easy-to-use APIs. You can use Amazon SQS and Amazon SNS to decouple and scale microservices, distributed systems, and serverless applications and improve overall reliability.

This blog looks at the first part of some fundamental integration patterns. We describe the patterns and apply them to these AWS messaging services. This will help you apply the right pattern to your use case and architect for scale in a secure and cost-efficient manner. For all variants, we employ both traditional and cloud-native messaging services: Amazon MQ for the former and Amazon SQS and Amazon SNS for the latter.

Integration Patterns

Let’s start with some fundamental integration patterns.

Message exchange patterns

First, we inspect the two major message exchange patterns: one-way and request-response.

One-way messaging

Applying one-way messaging, a message producer (sender) sends out a message to a messaging channel and doesn’t expect or want a response from whatever process (receiver) consumed the message. Examples of one-way messaging include a data transfer and a notification about an event that happened.

Request-response messaging

With request-response messaging, a message producer (requester) sends out a message: for example, a command to instruct the responder to execute something. The requester expects a response from each message consumer (responder) who received that message, likely to know what the result of all executions was. To know where to send the response message to, the request message contains a return address that the responder uses. To make sure that the requester can assign an incoming response to a request, the requester adds a correlation identifier to the request, which the responders echo in their responses.

Messaging channels: point-to-point

Next, we look at the point-to-point messaging channel, one of the most important patterns for messaging channels. We will continue our consideration with publish-subscribe in our second post.

A point-to-point channel is usually implemented by message queues. Message queues operate so that any given message is only consumed by one receiver, although multiple receivers can be connected to the queue. The queue ensures once-only consumption. Messages are usually buffered in queues so that they’re available for consumption for a certain amount of time, even if no receiver is currently connected.

Point-to-point channels are often used for loosely coupled message transmission, though there are two other common uses. First, it can support horizontal scaling of message processing on the receiver side. Depending on the message load in the channel, the number of receiver processes can be elastically adjusted to cope with the load as needed. The queue acts as a buffering load balancer. Second, it can flatten peak loads of messages and prevent your receivers from being flooded when you can’t scale out fast enough or you don’t want additional scaling.

Integration scenarios

In this section, we apply these fundamental patterns to AWS messaging services. The code examples are written in Java, but only by author preference. You can implement the same integration scenarios in C++, .NET, Node.js, Python, Ruby, Go, and other programming languages that AWS provides an SDK and an Apache Active MQ client library is available for.

Point-to-point channels: one-way messaging

The diagrams in the following subsections show the principle of one-way messaging for point-to-point channels, using Amazon MQ queues and Amazon SQS queues. The sender produces a message and sends it into a queue, and the receiver consumes the message from the queue for processing. For traditional messaging (that is, Amazon MQ), the senders and consumers can use protocols such as JMS or AMQP. For cloud-native messaging, they can use the Amazon SQS API.

Traditional messaging

To follow this example, open the Amazon MQ console and create a broker. In the following diagram we see the above explained components for the traditional messaging scenario: A sender sends messages into an Amazon MQ queue, a receiver consumes messages from that queue.

Point to point traditional messaging

In the following code example, sender and receiver are using the Apache Active MQ client library and the standard Java messaging service (JMS) API to send and receive messages to and from an Amazon MQ queue. You can run the code on every Amazon compute service, your on-premises data center, or your personal computer. For simplicity, the code launches sender and receiver in the same Java virtual machine (JVM).

public class PointToPointOneWayTraditional {

    public static void main(String... args) throws Exception {
        ActiveMQSslConnectionFactory connFact = new ActiveMQSslConnectionFactory("failover:(ssl://<broker-1>.amazonaws.com:61617,ssl://<broker-2>.amazonaws.com:61617)");
        Connection conn = connFact.createConnection("user", "password");

        new Thread(new Receiver(conn.createSession(false, Session.CLIENT_ACKNOWLEDGE), "Queue.PointToPoint.OneWay.Traditional")).start();
        new Thread(new Sender(conn.createSession(false, Session.CLIENT_ACKNOWLEDGE), "Queue.PointToPoint.OneWay.Traditional")).start();

    public static class Sender implements Runnable {

        private Session session;
        private String destination;

        public Sender(Session session, String destination) {
            this.session = session;
            this.destination = destination;

        public void run() {
            try {
                MessageProducer messageProducer = session.createProducer(session.createQueue(destination));
                long counter = 0;

                while (true) {
                    TextMessage message = session.createTextMessage("Message " + ++counter);
            } catch (JMSException e) {
                throw new RuntimeException(e);

    public static class Receiver implements Runnable, MessageListener {

        private Session session;
        private String destination;

        public Receiver(Session session, String destination) {
            this.session = session;
            this.destination = destination;

        public void run() {
            try {
                MessageConsumer consumer = session.createConsumer(session.createQueue(destination));
            } catch (JMSException e) {
                throw new RuntimeException(e);

        public void onMessage(Message message) {
            try {
                System.out.println(String.format("received message '%s' with message id '%s'", ((TextMessage) message).getText(), message.getJMSMessageID()));
            } catch (JMSException e) {
                throw new RuntimeException(e);

Cloud-native messaging

To follow this example, open the Amazon SQS console and create a standard SQS queue, using the queue name P2POneWayCloudNative.  In the following diagram we see the above explained components for the cloud-native messaging scenario: A sender sends messages into an Amazon SQS queue, a receiver consumes messages from that queue.

Point to point cloud-native messaging


In the sample code below, the example sender is using the AWS SDK for Java to send messages to an Amazon SQS queue, running in an endless loop. You can run the code on every Amazon compute service, your on-premises data center, or your personal computer.

public class PointToPointOneWayCloudNative {

    public static void main(String... args) throws Exception {
        final AmazonSQS sqs = AmazonSQSClientBuilder.standard().build();

        new Thread(new Sender(sqs, "https://sqs.<region>.amazonaws.com/<account-number>/P2POneWayCloudNative")).start();

    public static class Sender implements Runnable {

        private AmazonSQS sqs;
        private String destination;

        public Sender(AmazonSQS sqs, String destination) {
            this.sqs = sqs;
            this.destination = destination;

        public void run() {
            long counter = 0;

            while (true) {
                    new SendMessageRequest()
                        .withMessageBody("Message " + ++counter)
                        .addMessageAttributesEntry("MessageID", new MessageAttributeValue().withDataType("String").withStringValue(UUID.randomUUID().toString())));

We implement the receiver below in a serverless manner as an AWS Lambda function, using Amazon SQS as the event source. The name of the SQS queue is configured outside the function’s code, which is why it doesn’t appear in this code example.

public class Receiver implements RequestHandler<SQSEvent, Void> {

    public Void handleRequest(SQSEvent request, Context context) {
        for (SQSEvent.SQSMessage message: request.getRecords()) {
            System.out.println(String.format("received message '%s' with message id '%s'", message.getBody(), message.getMessageAttributes().get("MessageID").getStringValue()));

        return null;

If this approach is new to you, you can find more details in AWS Lambda Adds Amazon Simple Queue Service to Supported Event Sources. Using Lambda comes with a number of benefits. For example, you don’t have to manage the compute environment for the receiver, and you can use an event (or push) model instead of having to poll for new messages.

Point-to-point channels: request-response messaging

In addition to the one-way scenario, we have a return channel option. We would now call the involved processes rather than the requester and responder. The requester sends a message into the request queue, and the responder sends the response into the response queue. Remember that the requester enriches the message with a return address (the name of the response queue) so that the responder knows where to send the response to. The requester also sends a correlation ID that the responder copies into the response message so that the requester can match the incoming response with a request.

Traditional messaging

In this example, we reuse the Amazon MQ broker that we set up earlier. In the following diagram we see the above explained components for the traditional messaging scenario, using an Amazon MQ queue each for the request messages and for the response messages.

Point to point request response traditional messaging

Using Amazon MQ, we don’t have to create queues explicitly because they’re implicitly created as needed when we start sending messages to them. This example is similar to the point-to-point one-way traditional example.

public class PointToPointRequestResponseTraditional {

    public static void main(String... args) throws Exception {
        ActiveMQSslConnectionFactory connFact = new ActiveMQSslConnectionFactory("failover:(ssl://<broker-1>.amazonaws.com:61617,ssl://<broker-2>.amazonaws.com:61617)");
        Connection conn = connFact.createConnection("user", "password");

        new Thread(new Responder(conn.createSession(false, Session.CLIENT_ACKNOWLEDGE), "Queue.PointToPoint.RequestResponse.Traditional")).start();
        new Thread(new Requester(conn.createSession(false, Session.CLIENT_ACKNOWLEDGE), "Queue.PointToPoint.RequestResponse.Traditional")).start();

    public static class Requester implements Runnable {

        private Session session;
        private String destination;

        public Requester(Session session, String destination) {
            this.session = session;
            this.destination = destination;

        public void run() {
            MessageProducer messageProducer = null;
            try {
                messageProducer = session.createProducer(session.createQueue(destination));
                long counter = 0;

                while (true) {
                    TemporaryQueue replyTo = session.createTemporaryQueue();
                    String correlationId = UUID.randomUUID().toString();
                    TextMessage message = session.createTextMessage("Message " + ++counter);

                    MessageConsumer consumer = session.createConsumer(replyTo, "JMSCorrelationID='" + correlationId + "'");
                    try {
                        Message receivedMessage = consumer.receive(5000);
                        System.out.println(String.format("received message '%s' with message id '%s'", ((TextMessage) receivedMessage).getText(), receivedMessage.getJMSMessageID()));
                    } finally {
                        if (consumer != null) {
            } catch (JMSException e) {
                throw new RuntimeException(e);

    public static class Responder implements Runnable, MessageListener {

        private Session session;
        private String destination;

        public Responder(Session session, String destination) {
            this.session = session;
            this.destination = destination;

        public void run() {
            try {
                MessageConsumer consumer = session.createConsumer(session.createQueue(destination));
            } catch (JMSException e) {
                throw new RuntimeException(e);

        public void onMessage(Message message) {
            try {
                String correlationId = message.getJMSCorrelationID();
                Destination replyTo = message.getJMSReplyTo();

                TextMessage responseMessage = session.createTextMessage(((TextMessage) message).getText() + " with CorrelationID " + correlationId);

                MessageProducer messageProducer = session.createProducer(replyTo);
                try {

                } finally {
                    if (messageProducer != null) {
            } catch (JMSException e) {
                throw new RuntimeException(e);

Cloud-native messaging

Open the Amazon SQS console and create two standard SQS queues using the queue names P2PReqRespCloudNative and P2PReqRespCloudNative-Resp. In the following diagram we see the above explained components for the cloud-native scenario, using an Amazon SQS queue each for the request messages and for the response messages.

Point to point request response cloud native messaging

The following example requester is almost identical to the point-to-point one-way cloud-native example sender. It also provides a reply-to address and a correlation ID.

public class PointToPointRequestResponseCloudNative {

    public static void main(String... args) throws Exception {
        final AmazonSQS sqs = AmazonSQSClientBuilder.standard().build();

        new Thread(new Requester(sqs, "https://sqs.<region>.amazonaws.com/<account-number>/P2PReqRespCloudNative", "https://sqs.<region>.amazonaws.com/<account-number>/P2PReqRespCloudNative-Resp")).start();

    public static class Requester implements Runnable {

        private AmazonSQS sqs;
        private String destination;
        private String replyDestination;
        private Map<String, SendMessageRequest> inflightMessages = new ConcurrentHashMap<>();

        public Requester(AmazonSQS sqs, String destination, String replyDestination) {
            this.sqs = sqs;
            this.destination = destination;
            this.replyDestination = replyDestination;

        public void run() {
            long counter = 0;

            while (true) {
                String correlationId = UUID.randomUUID().toString();
                SendMessageRequest request = new SendMessageRequest()
                    .withMessageBody("Message " + ++counter)
                    .addMessageAttributesEntry("CorrelationID", new MessageAttributeValue().withDataType("String").withStringValue(correlationId))
                    .addMessageAttributesEntry("ReplyTo", new MessageAttributeValue().withDataType("String").withStringValue(replyDestination));

                inflightMessages.put(correlationId, request);

                ReceiveMessageResult receiveMessageResult = sqs.receiveMessage(
                    new ReceiveMessageRequest()

                for (Message receivedMessage : receiveMessageResult.getMessages()) {
                    System.out.println(String.format("received message '%s' with message id '%s'", receivedMessage.getBody(), receivedMessage.getMessageId()));

                    String receivedCorrelationId = receivedMessage.getMessageAttributes().get("CorrelationID").getStringValue();
                    SendMessageRequest originalRequest = inflightMessages.remove(receivedCorrelationId);
                    System.out.println(String.format("Corresponding request message '%s'", originalRequest.getMessageBody()));

                        new DeleteMessageRequest()

The following example responder is almost identical to the point-to-point one-way cloud-native example receiver. It also creates a message and sends it back to the reply-to address provided in the received message.

public class Responder implements RequestHandler<SQSEvent, Void> {

    private final AmazonSQS sqs = AmazonSQSClientBuilder.standard().build();

    public Void handleRequest(SQSEvent request, Context context) {
        for (SQSEvent.SQSMessage message: request.getRecords()) {
            System.out.println(String.format("received message '%s' with message id '%s'", message.getBody(), message.getMessageId()));
            String correlationId = message.getMessageAttributes().get("CorrelationID").getStringValue();
            String replyTo = message.getMessageAttributes().get("ReplyTo").getStringValue();

            System.out.println(String.format("sending message with correlation id '%s' to '%s'", correlationId, replyTo));
                new SendMessageRequest()
                    .withMessageBody(message.getBody() + " with CorrelationID " + correlationId)
                    .addMessageAttributesEntry("CorrelationID", new MessageAttributeValue().withDataType("String").withStringValue(correlationId)));

        return null;

Go build!

We look forward to hearing about what you build and will continue innovating our services on your behalf.

Additional resources

What’s next?

We have introduced the first fundamental EIPs and shown how you can apply them to the AWS messaging services. If you are keen to dive deeper, continue reading with the second part of this series, where we will cover publish-subscribe messaging.

Read Part 2: Publish-Subscribe Messaging

Encrypting messages published to Amazon SNS with AWS KMS

Post Syndicated from Michelle Mercier original https://aws.amazon.com/blogs/compute/encrypting-messages-published-to-amazon-sns-with-aws-kms/

Amazon Simple Notification Service (Amazon SNS) is a fully managed pub/sub messaging service for decoupling event-driven microservices, distributed systems, and serverless applications. Enterprises around the world use Amazon SNS to support applications that handle private and sensitive data. Many of these enterprises operate in regulated markets, and their systems are subject to stringent security and compliance standards, such as HIPAA for healthcare, PCI DSS for finance, and FedRAMP for government. To address the requirements of highly critical workloads, Amazon SNS provides message encryption in transit, based on Amazon Trust Services (ATS) certificates, as well as message encryption at rest, using AWS Key Management Service (AWS KMS) keys. 

Message encryption in transit

The Amazon SNS API is served through Secure HTTP (HTTPS) and encrypts all messages in transit with Transport Layer Security (TLS) certificates issued by ATS. These certificates verify the identity of the Amazon SNS API server whenever an encrypted connection is established. A certificate authority (CA) issues the certificate to a specific domain. Thus, when a domain presents a certificate issued by a trusted CA, the API client can determine that it’s safe to establish a connection.

Message encryption at rest

Amazon SNS supports encrypted topics. When you publish messages to encrypted topics, Amazon SNS uses customer master keys (CMK), powered by AWS KMS, to encrypt your messages. Amazon SNS supports customer-managed as well as AWS-managed CMKs. As soon as Amazon SNS receives your messages, the encryption takes place on the server, using a 256-bit AES-GCM algorithm. The messages are stored in encrypted form across multiple Availability Zones (AZs) for durability and are decrypted just before being delivered to subscribed endpoints, such as Amazon Simple Queue Service (Amazon SQS) queues, AWS Lambda functions, and HTTP and HTTPS webhooks.



  • Amazon SNS encrypts only the body of the messages that you publish. It doesn’t encrypt message metadata (identifier, subject, timestamp, and attributes), topic metadata (name and attributes), or topic metrics. Thus, encryption doesn’t affect the operations of Amazon SNS, such as message fanout and message filtering.
  • Amazon SNS doesn’t retroactively encrypt messages that were published before server-side encryption (SSE) was enabled for the topic. In addition, any encrypted message retains its encryption even if you disable the encryption of its topic. It can take up to a minute for encryption to be effective after enabled.
  • Amazon SNS uses envelope encryption internally. It uses your configured CMK to generate a short-lived data encryption key (DEK) and then reuses this DEK to encrypt your published messages for 5 minutes. When the DEK expires, Amazon SNS automatically rotates to generate a new DEK from AWS KMS.

Applying encrypted topics in a use case

You can use encrypted topics for a variety of scenarios, especially for processing sensitive data, such as personally identifiable information (PII) and protected health information (PHI).

The following example illustrates an electronic medical record (EMR) system deployed in several clinics and hospitals. The system manages patients’ medical histories, diagnoses, medications, immunizations, visits, lab results, and doctors’ notes.

The clinic’s EMR system is integrated with three auxiliary systems. Each system, hosted on an Amazon EC2 instance, polls incoming patient records from an Amazon SQS queue and takes action:

  • The Billing system manages the clinic’s revenue cycles and processes accounts receivable and reimbursements.
  • The Scheduling system keeps patients informed of their upcoming clinic and lab appointments and reminds them to take their medications.
  • The Prescription system transmits electronic prescriptions to authorized pharmacies and tracks medication-filling history.

When a physician inputs a new record into a patient’s file, the EMR system publishes a message to an Amazon SNS topic. The topic in turn fans out a copy of the message to each one of the three subscribing Amazon SQS queues for parallel processing. When the message is retrieved from the queue, the Billing system invoices the patient, the Scheduling system books the patient’s next clinic or lab appointment, and the Prescription system orders the required medication from an authorized pharmacy.

The Amazon SNS topic and all Amazon SQS queues described in this use case are encrypted using AWS KMS keys that the clinic creates. The communication among services is based on HTTPS API calls. This end-to-end encryption protects patients’ medical records by making their content unavailable to unauthorized or anonymous users while messages are either in transit or at rest.

Creating, subscribing, and publishing to encrypted topics

You can create an Amazon SNS encrypted topic or an Amazon SQS encrypted queue by setting its attribute KmsMasterKeyId, which expects an AWS KMS key identifier. The key identifier can be a key ID, key ARN, or key alias. You can use the identifier of either a customer-managed CMK, such as alias/MyKey, or the AWS-managed CMK in your account, whose alias is alias/aws/sns.

The following code snippets work with the AWS SDK for Java. You can use these code samples for the healthcare system scenario in the previous section.

First, the principal publishing messages to the Amazon SNS encrypted topic must have access permission to execute the AWS KMS operations GenerateDataKey and Decrypt, in addition to the Amazon SNS operation Publish. The principal can be either an IAM user or an IAM role. The following IAM policy grants the required access permission to the principal.

  "Version": "2012-10-17",
  "Statement": {
    "Effect": "Allow",
    "Action": [
    "Resource": "*"

If you want to use a customer-managed CMK, a CMK needs to be created and secured by granting the publisher access to the same AWS KMS operations GenerateDataKey and Decrypt. The access permission is granted using KMS key policies. The following JSON document shows an example policy statement for the customer-managed CMK used by the healthcare system. For more information on creating and securing keys, see Creating Keys and Using Key Policies in the AWS Key Management Service Developer Guide.

  "Version": "2012-10-17",
  "Id": "EMR-System-KeyPolicy",
  "Statement": [
      "Sid": "Allow access for Root User",
      "Effect": "Allow",
      "Principal": {"AWS": "arn:aws:iam::123456789012:root"},
      "Action": "kms:*",
      "Resource": "*"
      "Sid": "Allow access for Key Administrator",
      "Effect": "Allow",
      "Principal": {"AWS": "arn:aws:iam::123456789012:user/EMRAdmin"},
      "Action": [
      "Resource": "*"
      "Sid": "Allow access for Key User (SNS Publisher)",
      "Effect": "Allow",
      "Principal": {"AWS": "arn:aws:iam::123456789012:user/EMRUser"},
      "Action": [
      "Resource": "*"

The following snippet uses the CMK to create an encrypted topic, three encrypted queues, and their corresponding subscriptions.

// Create API clients

String userArn = "arn:aws:iam::123456789012:user/EMRUser";

AWSCredentialsProvider credentials = getCredentials(userArn);

AmazonSNS sns = new AmazonSNSClient(credentials);
AmazonSQS sqs = new AmazonSQSClient(credentials);

// Create an attributes collection for the topic and queues

String keyId = "arn:aws:kms:us-east-1:123456789012:alias/EMRKey"; 

Map<String, String> attributes = new HashMap<>();
attributes.put("KmsMasterKeyId", keyId);

// Create an encrypted topic

String topicArn = sns.createTopic(
    new CreateTopicRequest("Patient-Records")

// Create encrypted queues

String billingQueueUrl = sqs.createQueue(
    new CreateQueueRequest("Billing-Integration")

String schedulingQueueUrl = sqs.createQueue(
    new CreateQueueRequest("Scheduling-Integration")

String prescriptionQueueUrl = sqs.createQueue(
    new CreateQueueRequest("Prescription-Integration")

// Create subscriptions

Topics.subscribeQueue(sns, sqs, topicArn, billingQueueUrl);
Topics.subscribeQueue(sns, sqs, topicArn, schedulingQueueUrl);
Topics.subscribeQueue(sns, sqs, topicArn, prescriptionQueueUrl);

Next, the following code composes a JSON message and publishes it to the encrypted topic.

// Publish message to encrypted topic

String messageBody = "{\"patient\": 2911, \"medication\": 151}";
String messageSubject = "Electronic Medical Record - 3472";

    new PublishRequest()


Publishing messages to encrypted topics isn’t different from publishing messages to standard, unencrypted topics. Your publisher needs access to perform AWS KMS operations GenerateDataKey and Decrypt on the configured CMK. All of the encryption logic is offloaded to Amazon SNS, and the message is delivered to all subscribed endpoints.

A copy of the message is now available in each subscribing queue. The final code snippet retrieves the messages from the encrypted queues.

// Retrieve messages from encrypted queues

List<Message> messagesForBilling = sqs.receiveMessage(
    new ReceiveMessageRequest(billingQueueUrl)).getMessages();

List<Message> messagesForScheduling = sqs.receiveMessage(
    new ReceiveMessageRequest(schedulingQueueUrl)).getMessages();

List<Message> messagesForPrescription = sqs.receiveMessage(
    new ReceiveMessageRequest(prescriptionQueueUrl)).getMessages();


Retrieving messages from encrypted queues isn’t different from retrieving messages from standard, unencrypted queues. All of the decryption logic is offloaded to Amazon SQS.

Enabling compatibility between encrypted topics and event sources

Several AWS services publish events to Amazon SNS topics. To allow these event sources to work with encrypted topics, you must first create a customer-managed CMK and then add the following statement to the policy of the CMK.

    "Version": "2012-10-17",
    "Statement": [{
        "Effect": "Allow",
        "Principal": {"Service": "service.amazonaws.com"},
        "Action": ["kms:GenerateDataKey*", "kms:Decrypt"],
        "Resource": "*"

You can use the following example service principals in the statement:

Other Amazon SNS event sources require you to provide an IAM role, as opposed to their service principal, in the KMS key policy. This set of event sources includes the following:

Once the CMK key policy has been configured, you can enable encryption on the topic using the CMK, and then provide the encrypted topic’s ARN to the event source.


As of November 2018, Amazon CloudWatch alarms don’t yet work with Amazon SNS encrypted topics. For information on publishing alarms to standard, unencrypted topics, see Using Amazon CloudWatch Alarms in the Amazon CloudWatch User Guide.

Publishing messages privately to encrypted topics through VPC endpoints

In addition to encrypting messages in transit and at rest, you can further harden the security and privacy of your applications by publishing messages to encrypted topics privately, without traversing the public Internet. Amazon SNS supports VPC endpoints via AWS PrivateLink. You can use VPC endpoints to privately publish messages to both standard and encrypted topics from a virtual private cloud (VPC) subnet. When you use AWS PrivateLink, you don’t have to set up an internet gateway, network address translation (NAT) device, or virtual private network (VPN) connection. For more information, see Publishing to Amazon SNS topics from Amazon Virtual Private Cloud in the Amazon Simple Notification Service Developer Guide.

Auditing the usage of encrypted topics

You can use AWS CloudTrail to audit the usage of the AWS KMS keys applied to your Amazon SNS topics. AWS CloudTrail creates log files that contain a history of AWS API calls and related events for your account. These log files include all AWS KMS API requests made with the AWS Management Console , SDKs, and Command Line Tools, as well as those made through integrated AWS services. You can use these log files to get information about when your CMK was used, the operation that was requested, the identity of the requester, and the IP address that the request came from. For more information, see Logging AWS KMS API calls with AWS CloudTrail in the AWS Key Management Service Developer Guide.


Amazon SNS provides a full set of security features to protect your data from unauthorized and anonymous access, including message encryption in transit with Amazon ATS certificates, message encryption at rest with AWS KMS keys, message privacy with AWS PrivateLink, and auditing with AWS CloudTrail. Moreover, you can subscribe Amazon SQS encrypted queues to Amazon SNS encrypted topics to establish end-to-end encryption in your messaging use cases.

Amazon SNS encrypted topics are available in all AWS Regions where AWS KMS is available. For pricing details, see AWS KMS pricing and Amazon SNS pricing. There is no increase in Amazon SNS charges for using encrypted topics, beyond the AWS KMS request charges incurred. For more information, see Protecting Amazon SNS Data Using Server-Side Encryption (SSE) in the Amazon Simple Notification Service Developer Guide.

Get started today by creating your Amazon SNS encrypted topics via the AWS Management Console and AWS SDKs.

Measuring the throughput for Amazon MQ using the JMS Benchmark

Post Syndicated from Rachel Richardson original https://aws.amazon.com/blogs/compute/measuring-the-throughput-for-amazon-mq-using-the-jms-benchmark/

This post is courtesy of Alan Protasio, Software Development Engineer, Amazon Web Services

Just like compute and storage, messaging is a fundamental building block of enterprise applications. Message brokers (aka “message-oriented middleware”) enable different software systems, often written in different languages, on different platforms, running in different locations, to communicate and exchange information. Mission-critical applications, such as CRM and ERP, rely on message brokers to work.

A common performance consideration for customers deploying a message broker in a production environment is the throughput of the system, measured as messages per second. This is important to know so that application environments (hosts, threads, memory, etc.) can be configured correctly.

In this post, we demonstrate how to measure the throughput for Amazon MQ, a new managed message broker service for ActiveMQ, using JMS Benchmark. It should take between 15–20 minutes to set up the environment and an hour to run the benchmark. We also provide some tips on how to configure Amazon MQ for optimal throughput.

Benchmarking throughput for Amazon MQ

ActiveMQ can be used for a number of use cases. These use cases can range from simple fire and forget tasks (that is, asynchronous processing), low-latency request-reply patterns, to buffering requests before they are persisted to a database.

The throughput of Amazon MQ is largely dependent on the use case. For example, if you have non-critical workloads such as gathering click events for a non-business-critical portal, you can use ActiveMQ in a non-persistent mode and get extremely high throughput with Amazon MQ.

On the flip side, if you have a critical workload where durability is extremely important (meaning that you can’t lose a message), then you are bound by the I/O capacity of your underlying persistence store. We recommend using mq.m4.large for the best results. The mq.t2.micro instance type is intended for product evaluation. Performance is limited, due to the lower memory and burstable CPU performance.

Tip: To improve your throughput with Amazon MQ, make sure that you have consumers processing messaging as fast as (or faster than) your producers are pushing messages.

Because it’s impossible to talk about how the broker (ActiveMQ) behaves for each and every use case, we walk through how to set up your own benchmark for Amazon MQ using our favorite open-source benchmarking tool: JMS Benchmark. We are fans of the JMS Benchmark suite because it’s easy to set up and deploy, and comes with a built-in visualizer of the results.

Non-Persistent Scenarios – Queue latency as you scale producer throughput

JMS Benchmark nonpersistent scenarios

Getting started

At the time of publication, you can create an mq.m4.large single-instance broker for testing for $0.30 per hour (US pricing).

This walkthrough covers the following tasks:

  1.  Create and configure the broker.
  2. Create an EC2 instance to run your benchmark
  3. Configure the security groups
  4.  Run the benchmark.

Step 1 – Create and configure the broker
Create and configure the broker using Tutorial: Creating and Configuring an Amazon MQ Broker.

Step 2 – Create an EC2 instance to run your benchmark
Launch the EC2 instance using Step 1: Launch an Instance. We recommend choosing the m5.large instance type.

Step 3 – Configure the security groups
Make sure that all the security groups are correctly configured to let the traffic flow between the EC2 instance and your broker.

  1. Sign in to the Amazon MQ console.
  2. From the broker list, choose the name of your broker (for example, MyBroker)
  3. In the Details section, under Security and network, choose the name of your security group or choose the expand icon ( ).
  4. From the security group list, choose your security group.
  5. At the bottom of the page, choose Inbound, Edit.
  6. In the Edit inbound rules dialog box, add a role to allow traffic between your instance and the broker:
    • Choose Add Rule.
    • For Type, choose Custom TCP.
    • For Port Range, type the ActiveMQ SSL port (61617).
    • For Source, leave Custom selected and then type the security group of your EC2 instance.
    • Choose Save.

Your broker can now accept the connection from your EC2 instance.

Step 4 – Run the benchmark
Connect to your EC2 instance using SSH and run the following commands:

$ cd ~
$ curl -L https://github.com/alanprot/jms-benchmark/archive/master.zip -o master.zip
$ unzip master.zip
$ cd jms-benchmark-master
$ chmod a+x bin/*
$ env \
  SERVER_SETUP=false \
  SERVER_ADDRESS={activemq-endpoint} \
  ACTIVEMQ_USERNAME={activemq-user} \
  ACTIVEMQ_PASSWORD={activemq-password} \

After the benchmark finishes, you can find the results in the ~/reports directory. As you may notice, the performance of ActiveMQ varies based on the number of consumers, producers, destinations, and message size.

Amazon MQ architecture

The last bit that’s important to know so that you can better understand the results of the benchmark is how Amazon MQ is architected.

Amazon MQ is architected to be highly available (HA) and durable. For HA, we recommend using the multi-AZ option. After a message is sent to Amazon MQ in persistent mode, the message is written to the highly durable message store that replicates the data across multiple nodes in multiple Availability Zones. Because of this replication, for some use cases you may see a reduction in throughput as you migrate to Amazon MQ. Customers have told us they appreciate the benefits of message replication as it helps protect durability even in the face of the loss of an Availability Zone.


We hope this gives you an idea of how Amazon MQ performs. We encourage you to run tests to simulate your own use cases.

To learn more, see the Amazon MQ website. You can try Amazon MQ for free with the AWS Free Tier, which includes up to 750 hours of a single-instance mq.t2.micro broker and up to 1 GB of storage per month for one year.

Solving Complex Ordering Challenges with Amazon SQS FIFO Queues

Post Syndicated from Christie Gifrin original https://aws.amazon.com/blogs/compute/solving-complex-ordering-challenges-with-amazon-sqs-fifo-queues/

Contributed by Shea Lutton, AWS Cloud Infrastructure Architect

Amazon Simple Queue Service (Amazon SQS) is a fully managed queuing service that helps decouple applications, distributed systems, and microservices to increase fault tolerance. SQS queues come in two distinct types:

  • Standard SQS queues are able to scale to enormous throughput with at-least-once delivery.
  • FIFO queues are designed to guarantee that messages are processed exactly once in the exact order that they are received and have a default rate of 300 transactions per second.

As customers explore SQS FIFO queues, they often have questions about how the behavior works when messages arrive and are consumed. This post walks through some common situations to identify the exact behavior that you can expect. It also covers the behavior of message groups in depth and explains why message groups are key to understanding how FIFO queues work.

The simple case

Suppose that you run a major auction platform where people buy and sell a wide range of products. Your platform requires that transactions from buyers and sellers get processed in exactly the order received. Here’s how a FIFO queue helps you keep all your transactions in one straight flow.

A seller currently is holding an auction for a laptop, and three different bids are received for the same price. Ties are awarded to the first bidder at that price so it is important to track which arrived first. Your auction platform receives the three bids and sends them to a FIFO queue before they are processed.

Now observe how messages leave the queue. When your consumer asks for a batch of up to 10 messages, SQS starts filling the batch with the oldest message (bid A1). It keeps filling until either the batch is full or the queue is empty. In this case, the batch contains the three messages and the queue is now empty. After a batch has left the queue, SQS considers that batch of messages to be “in-flight” until the consumer either deletes them or the batch’s visibility timer expires.


When you have a single consumer, this is easy to envision. The consumer gets a batch of messages (now in-flight), does its processing, and deletes the messages. That consumer is then ready to ask for the next batch of messages.

The critical thing to keep in mind is that SQS won’t release the next batch of messages until the first batch has been deleted. By adding more messages to the queue, you can see more interesting behaviors. Imagine that a burst of 11 bids is sent to your FIFO queue, with two bids for Auction A arriving last.

The FIFO queue now has at least two batches of messages in it. When your single consumer requests the first batch of 10 messages, it receives a batch starting with B1 and ending with A1. Later, after the first batch has been deleted, the consumer can get the second batch of messages containing the final A2 message from the queue.

Adding complexity with multiple message groups

A new challenge arises. Your auction platform is getting busier and your dev team added a number of new features. The combination of increased messages and extra processing time for the new features means that a single consumer is too slow. The solution is to scale to have more consumers and process messages in parallel.

To work in parallel, your team realized that only the messages related to a single auction must be kept in order. All transactions for Auction A need to be kept in order and so do all transactions for Auction B. But the two auctions are independent and it does not matter which auctions transactions are processed first.

FIFO can handle that case with a feature called message groups. Each transaction related to Auction A is placed by your producer into message group A, and so on. In the diagram below, Auction A and Auction B each received three bid transactions, with bid B1 arriving first. The FIFO queue always keeps transactions within a message group in the order in which they arrived.

How is this any different than earlier examples? The consumer now gets the messages ordered by message groups, all the B group messages followed by all the A group messages. Multiple message groups create the possibility of using multiple consumers, which I explain in a moment. If FIFO can’t fill up a batch of messages with a single message group, FIFO can place more than one message group in a batch of messages. But whenever possible, the queue gives you a full batch of messages from the same group.

The order of messages leaving a FIFO queue is governed by three rules:

  1. Return the oldest message where no other message in the same message group is currently in-flight.
  2. Return as many messages from the same message group as possible.
  3. If a message batch is still not full, go back to rule 1.

To see this behavior, add a second consumer and insert many more messages into the queue. For simplicity, the delete message action has been omitted in these diagrams but it is assumed that all messages in a batch are processed successfully by the consumer and the batch is properly deleted immediately after.

In this example, there are 11 Group A and 11 Group B transactions arriving in interleaved order and a second consumer has been added. Consumer 1 asks for a group of 10 messages and receives 10 Group A messages. Consumer 2 then asks for 10 messages but SQS knows that Group A is in flight, so it releases 10 Group B messages. The two consumers are now processing two batches of messages in parallel, speeding up throughput and then deleting their batches. When Consumer 1 requests the next batch of messages, it receives the remaining two messages, one from Group A and one from Group B.

Consider this nuanced detail from the example above. What would happen if Consumer 1 was on a faster server and processed its first batch of messages before Consumer 2 could mark its messages for deletion? See if you can predict the behavior before looking at the answer.

If Consumer 2 has not deleted its Group B messages yet when Consumer 1 asks for the next batch, then the FIFO queue considers Group B to still be in flight. It does not release any more Group B messages. Consumer 1 gets only the remaining Group A message. Later, after Consumer 2 has deleted its first batch, the remaining Group B message is released.


I hope this post answered your questions about how Amazon SQS FIFO queues work and why message groups are helpful. If you’re interested in exploring SQS FIFO queues further, here are a few ideas to get you started:

AWS Online Tech Talks – May and Early June 2018

Post Syndicated from Devin Watson original https://aws.amazon.com/blogs/aws/aws-online-tech-talks-may-and-early-june-2018/

AWS Online Tech Talks – May and Early June 2018  

Join us this month to learn about some of the exciting new services and solution best practices at AWS. We also have our first re:Invent 2018 webinar series, “How to re:Invent”. Sign up now to learn more, we look forward to seeing you.

Note – All sessions are free and in Pacific Time.

Tech talks featured this month:

Analytics & Big Data

May 21, 2018 | 11:00 AM – 11:45 AM PT Integrating Amazon Elasticsearch with your DevOps Tooling – Learn how you can easily integrate Amazon Elasticsearch Service into your DevOps tooling and gain valuable insight from your log data.

May 23, 2018 | 11:00 AM – 11:45 AM PTData Warehousing and Data Lake Analytics, Together – Learn how to query data across your data warehouse and data lake without moving data.

May 24, 2018 | 11:00 AM – 11:45 AM PTData Transformation Patterns in AWS – Discover how to perform common data transformations on the AWS Data Lake.


May 29, 2018 | 01:00 PM – 01:45 PM PT – Creating and Managing a WordPress Website with Amazon Lightsail – Learn about Amazon Lightsail and how you can create, run and manage your WordPress websites with Amazon’s simple compute platform.

May 30, 2018 | 01:00 PM – 01:45 PM PTAccelerating Life Sciences with HPC on AWS – Learn how you can accelerate your Life Sciences research workloads by harnessing the power of high performance computing on AWS.


May 24, 2018 | 01:00 PM – 01:45 PM PT – Building Microservices with the 12 Factor App Pattern on AWS – Learn best practices for building containerized microservices on AWS, and how traditional software design patterns evolve in the context of containers.


May 21, 2018 | 01:00 PM – 01:45 PM PTHow to Migrate from Cassandra to Amazon DynamoDB – Get the benefits, best practices and guides on how to migrate your Cassandra databases to Amazon DynamoDB.

May 23, 2018 | 01:00 PM – 01:45 PM PT5 Hacks for Optimizing MySQL in the Cloud – Learn how to optimize your MySQL databases for high availability, performance, and disaster resilience using RDS.


May 23, 2018 | 09:00 AM – 09:45 AM PT.NET Serverless Development on AWS – Learn how to build a modern serverless application in .NET Core 2.0.

Enterprise & Hybrid

May 22, 2018 | 11:00 AM – 11:45 AM PTHybrid Cloud Customer Use Cases on AWS – Learn how customers are leveraging AWS hybrid cloud capabilities to easily extend their datacenter capacity, deliver new services and applications, and ensure business continuity and disaster recovery.


May 31, 2018 | 11:00 AM – 11:45 AM PTUsing AWS IoT for Industrial Applications – Discover how you can quickly onboard your fleet of connected devices, keep them secure, and build predictive analytics with AWS IoT.

Machine Learning

May 22, 2018 | 09:00 AM – 09:45 AM PTUsing Apache Spark with Amazon SageMaker – Discover how to use Apache Spark with Amazon SageMaker for training jobs and application integration.

May 24, 2018 | 09:00 AM – 09:45 AM PTIntroducing AWS DeepLens – Learn how AWS DeepLens provides a new way for developers to learn machine learning by pairing the physical device with a broad set of tutorials, examples, source code, and integration with familiar AWS services.

Management Tools

May 21, 2018 | 09:00 AM – 09:45 AM PTGaining Better Observability of Your VMs with Amazon CloudWatch – Learn how CloudWatch Agent makes it easy for customers like Rackspace to monitor their VMs.


May 29, 2018 | 11:00 AM – 11:45 AM PT – Deep Dive on Amazon Pinpoint Segmentation and Endpoint Management – See how segmentation and endpoint management with Amazon Pinpoint can help you target the right audience.


May 31, 2018 | 09:00 AM – 09:45 AM PTMaking Private Connectivity the New Norm via AWS PrivateLink – See how PrivateLink enables service owners to offer private endpoints to customers outside their company.

Security, Identity, & Compliance

May 30, 2018 | 09:00 AM – 09:45 AM PT – Introducing AWS Certificate Manager Private Certificate Authority (CA) – Learn how AWS Certificate Manager (ACM) Private Certificate Authority (CA), a managed private CA service, helps you easily and securely manage the lifecycle of your private certificates.

June 1, 2018 | 09:00 AM – 09:45 AM PTIntroducing AWS Firewall Manager – Centrally configure and manage AWS WAF rules across your accounts and applications.


May 22, 2018 | 01:00 PM – 01:45 PM PTBuilding API-Driven Microservices with Amazon API Gateway – Learn how to build a secure, scalable API for your application in our tech talk about API-driven microservices.


May 30, 2018 | 11:00 AM – 11:45 AM PTAccelerate Productivity by Computing at the Edge – Learn how AWS Snowball Edge support for compute instances helps accelerate data transfers, execute custom applications, and reduce overall storage costs.

June 1, 2018 | 11:00 AM – 11:45 AM PTLearn to Build a Cloud-Scale Website Powered by Amazon EFS – Technical deep dive where you’ll learn tips and tricks for integrating WordPress, Drupal and Magento with Amazon EFS.





Running ActiveMQ in a Hybrid Cloud Environment with Amazon MQ

Post Syndicated from Tara Van Unen original https://aws.amazon.com/blogs/compute/running-activemq-in-a-hybrid-cloud-environment-with-amazon-mq/

This post courtesy of Greg Share, AWS Solutions Architect

Many organizations, particularly enterprises, rely on message brokers to connect and coordinate different systems. Message brokers enable distributed applications to communicate with one another, serving as the technological backbone for their IT environment, and ultimately their business services. Applications depend on messaging to work.

In many cases, those organizations have started to build new or “lift and shift” applications to AWS. In some cases, there are applications, such as mainframe systems, too costly to migrate. In these scenarios, those on-premises applications still need to interact with cloud-based components.

Amazon MQ is a managed message broker service for ActiveMQ that enables organizations to send messages between applications in the cloud and on-premises to enable hybrid environments and application modernization. For example, you can invoke AWS Lambda from queues and topics managed by Amazon MQ brokers to integrate legacy systems with serverless architectures. ActiveMQ is an open-source message broker written in Java that is packaged with clients in multiple languages, Java Message Server (JMS) client being one example.

This post shows you can use Amazon MQ to integrate on-premises and cloud environments using the network of brokers feature of ActiveMQ. It provides configuration parameters for a one-way duplex connection for the flow of messages from an on-premises ActiveMQ message broker to Amazon MQ.

ActiveMQ and the network of brokers

First, look at queues within ActiveMQ and then at the network of brokers as a mechanism to distribute messages.

The network of brokers behaves differently from models such as physical networks. The key consideration is that the production (sending) of a message is disconnected from the consumption of that message. Think of the delivery of a parcel: The parcel is sent by the supplier (producer) to the end customer (consumer). The path it took to get there is of little concern to the customer, as long as it receives the package.

The same logic can be applied to the network of brokers. Here’s how you build the flow from a simple message to a queue and build toward a network of brokers. Before you look at setting up a hybrid connection, I discuss how a broker processes messages in a simple scenario.

When a message is sent from a producer to a queue on a broker, the following steps occur:

  1. A message is sent to a queue from the producer.
  2. The broker persists this in its store or journal.
  3. At this point, an acknowledgement (ACK) is sent to the producer from the broker.

When a consumer looks to consume the message from that same queue, the following steps occur:

  1. The message listener (consumer) calls the broker, which creates a subscription to the queue.
  2. Messages are fetched from the message store and sent to the consumer.
  3. The consumer acknowledges that the message has been received before processing it.
  4. Upon receiving the ACK, the broker sets the message as having been consumed. By default, this deletes it from the queue.
    • You can set the consumer to ACK after processing by setting up transaction management or handle it manually using Session.CLIENT_ACKNOWLEDGE.

Static propagation

I now introduce the concept of static propagation with the network of brokers as the mechanism for message transfer from on-premises brokers to Amazon MQ.  Static propagation refers to message propagation that occurs in the absence of subscription information. In this case, the objective is to transfer messages arriving at your selected on-premises broker to the Amazon MQ broker for consumption within the cloud environment.

After you configure static propagation with a network of brokers, the following occurs:

  1. The on-premises broker receives a message from a producer for a specific queue.
  2. The on-premises broker sends (statically propagates) the message to the Amazon MQ broker.
  3. The Amazon MQ broker sends an acknowledgement to the on-premises broker, which marks the message as having been consumed.
  4. Amazon MQ holds the message in its queue ready for consumption.
  5. A consumer connects to Amazon MQ broker, subscribes to the queue in which the message resides, and receives the message.
  6. Amazon MQ broker marks the message as having been consumed.

Getting started

The first step is creating an Amazon MQ broker.

  1. Sign in to the Amazon MQ console and launch a new Amazon MQ broker.
  2. Name your broker and choose Next step.
  3. For Broker instance type, choose your instance size:
  4. For Deployment mode, enter one of the following:
    Single-instance broker for development and test implementations (recommended)
    Active/standby broker for high availability in production environments
  5. Scroll down and enter your user name and password.
  6. Expand Advanced Settings.
  7. For VPC, Subnet, and Security Group, pick the values for the resources in which your broker will reside.
  8. For Public Accessibility, choose Yes, as connectivity is internet-based. Another option would be to use private connectivity between your on-premises network and the VPC, an example being an AWS Direct Connect or VPN connection. In that case, you could set Public Accessibility to No.
  9. For Maintenance, leave the default value, No preference.
  10. Choose Create Broker. Wait several minutes for the broker to be created.

After creation is complete, you see your broker listed.

For connectivity to work, you must configure the security group where Amazon MQ resides. For this post, I focus on the OpenWire protocol.

For Openwire connectivity, allow port 61617 access for Amazon MQ from your on-premises ActiveMQ broker source IP address. For alternate protocols, see the Amazon MQ broker configuration information for the ports required:

OpenWire – ssl://xxxxxxx.xxx.com:61617
AMQP – amqp+ssl:// xxxxxxx.xxx.com:5671
STOMP – stomp+ssl:// xxxxxxx.xxx.com:61614
MQTT – mqtt+ssl:// xxxxxxx.xxx.com:8883
WSS – wss:// xxxxxxx.xxx.com:61619

Configuring the network of brokers

Configuring the network of brokers with static propagation occurs on the on-premises broker by applying changes to the following file:
<activemq install directory>/conf activemq.xml

Network connector

This is the first configuration item required to enable a network of brokers. It is only required on the on-premises broker, which initiates and creates the connection with Amazon MQ. This connection, after it’s established, enables the flow of messages in either direction between the on-premises broker and Amazon MQ. The focus of this post is the uni-directional flow of messages from the on-premises broker to Amazon MQ.

The default activemq.xml file does not include the network connector configuration. Add this with the networkConnector element. In this scenario, edit the on-premises broker activemq.xml file to include the following information between <systemUsage> and <transportConnectors>:

                name="Q:source broker name->target broker name"
                uri="static:(ssl:// aws mq endpoint:61617)" 
                    <queue physicalName="queuename"/>
                      <queue physicalName=">" />

The highlighted components are the most important elements when configuring your on-premises broker.

  • name – Name of the network bridge. In this case, it specifies two things:
    • That this connection relates to an ActiveMQ queue (Q) as opposed to a topic (T), for reference purposes.
    • The source broker and target broker.
  • duplex –Setting this to false ensures that messages traverse uni-directionally from the on-premises broker to Amazon MQ.
  • uri –Specifies the remote endpoint to which to connect for message transfer. In this case, it is an Openwire endpoint on your Amazon MQ broker. This information could be obtained from the Amazon MQ console or via the API.
  • username and password – The same username and password configured when creating the Amazon MQ broker, and used to access the Amazon MQ ActiveMQ console.
  • networkTTL – Number of brokers in the network through which messages and subscriptions can pass. Leave this setting at the current value, if it is already included in your broker connection.
  • staticallyIncludedDestinations > queue physicalName – The destination ActiveMQ queue for which messages are destined. This is the queue that is propagated from the on-premises broker to the Amazon MQ broker for message consumption.

After the network connector is configured, you must restart the ActiveMQ service on the on-premises broker for the changes to be applied.

Verify the configuration

There are a number of places within the ActiveMQ console of your on-premises and Amazon MQ brokers to browse to verify that the configuration is correct and the connection has been established.

On-premises broker

Launch the ActiveMQ console of your on-premises broker and navigate to Network. You should see an active network bridge similar to the following:

This identifies that the connection between your on-premises broker and your Amazon MQ broker is up and running.

Now navigate to Connections and scroll to the bottom of the page. Under the Network Connectors subsection, you should see a connector labeled with the name: value that you provided within the ActiveMQ.xml configuration file. You should see an entry similar to:

Amazon MQ broker

Launch the ActiveMQ console of your Amazon MQ broker and navigate to Connections. Scroll to the Connections openwire subsection and you should see a connection specified that references the name: value that you provided within the ActiveMQ.xml configuration file. You should see an entry similar to:

If you configured the uri: for AMQP, STOMP, MQTT, or WSS as opposed to Openwire, you would see this connection under the corresponding section of the Connections page.

Testing your message flow

The setup described outlines a way for messages produced on premises to be propagated to the cloud for consumption in the cloud. This section provides steps on verifying the message flow.

Verify that the queue has been created

After you specify this queue name as staticallyIncludedDestinations > queue physicalName: and your ActiveMQ service starts, you see the following on your on-premises ActiveMQ console Queues page.

As you can see, no messages have been sent but you have one consumer listed. If you then choose Active Consumers under the Views column, you see Active Consumers for TestingQ.

This is telling you that your Amazon MQ broker is a consumer of your on-premises broker for the testing queue.

Produce and send a message to the on-premises broker

Now, produce a message on an on-premises producer and send it to your on-premises broker to a queue named TestingQ. If you navigate back to the queues page of your on-premises ActiveMQ console, you see that the messages enqueued and messages dequeued column count for your TestingQ queue have changed:

What this means is that the message originating from the on-premises producer has traversed the on-premises broker and propagated immediately to the Amazon MQ broker. At this point, the message is no longer available for consumption from the on-premises broker.

If you access the ActiveMQ console of your Amazon MQ broker and navigate to the Queues page, you see the following for the TestingQ queue:

This means that the message originally sent to your on-premises broker has traversed the network of brokers unidirectional network bridge, and is ready to be consumed from your Amazon MQ broker. The indicator is the Number of Pending Messages column.

Consume the message from an Amazon MQ broker

Connect to the Amazon MQ TestingQ queue from a consumer within the AWS Cloud environment for message consumption. Log on to the ActiveMQ console of your Amazon MQ broker and navigate to the Queue page:

As you can see, the Number of Pending Messages column figure has changed to 0 as that message has been consumed.

This diagram outlines the message lifecycle from the on-premises producer to the on-premises broker, traversing the hybrid connection between the on-premises broker and Amazon MQ, and finally consumption within the AWS Cloud.


This post focused on an ActiveMQ-specific scenario for transferring messages within an ActiveMQ queue from an on-premises broker to Amazon MQ.

For other on-premises brokers, such as IBM MQ, another approach would be to run ActiveMQ on-premises broker and use JMS bridging to IBM MQ, while using the approach in this post to forward to Amazon MQ. Yet another approach would be to use Apache Camel for more sophisticated routing.

I hope that you have found this example of hybrid messaging between an on-premises environment in the AWS Cloud to be useful. Many customers are already using on-premises ActiveMQ brokers, and this is a great use case to enable hybrid cloud scenarios.

To learn more, see the Amazon MQ website and Developer Guide. You can try Amazon MQ for free with the AWS Free Tier, which includes up to 750 hours of a single-instance mq.t2.micro broker and up to 1 GB of storage per month for one year.


Invoking AWS Lambda from Amazon MQ

Post Syndicated from Tara Van Unen original https://aws.amazon.com/blogs/compute/invoking-aws-lambda-from-amazon-mq/

Contributed by Josh Kahn, AWS Solutions Architect

Message brokers can be used to solve a number of needs in enterprise architectures, including managing workload queues and broadcasting messages to a number of subscribers. Amazon MQ is a managed message broker service for Apache ActiveMQ that makes it easy to set up and operate message brokers in the cloud.

In this post, I discuss one approach to invoking AWS Lambda from queues and topics managed by Amazon MQ brokers. This and other similar patterns can be useful in integrating legacy systems with serverless architectures. You could also integrate systems already migrated to the cloud that use common APIs such as JMS.

For example, imagine that you work for a company that produces training videos and which recently migrated its video management system to AWS. The on-premises system used to publish a message to an ActiveMQ broker when a video was ready for processing by an on-premises transcoder. However, on AWS, your company uses Amazon Elastic Transcoder. Instead of modifying the management system, Lambda polls the broker for new messages and starts a new Elastic Transcoder job. This approach avoids changes to the existing application while refactoring the workload to leverage cloud-native components.

This solution uses Amazon CloudWatch Events to trigger a Lambda function that polls the Amazon MQ broker for messages. Instead of starting an Elastic Transcoder job, the sample writes the received message to an Amazon DynamoDB table with a time stamp indicating the time received.

Getting started

To start, navigate to the Amazon MQ console. Next, launch a new Amazon MQ instance, selecting Single-instance Broker and supplying a broker name, user name, and password. Be sure to document the user name and password for later.

For the purposes of this sample, choose the default options in the Advanced settings section. Your new broker is deployed to the default VPC in the selected AWS Region with the default security group. For this post, you update the security group to allow access for your sample Lambda function. In a production scenario, I recommend deploying both the Lambda function and your Amazon MQ broker in your own VPC.

After several minutes, your instance changes status from “Creation Pending” to “Available.” You can then visit the Details page of your broker to retrieve connection information, including a link to the ActiveMQ web console where you can monitor the status of your broker, publish test messages, and so on. In this example, use the Stomp protocol to connect to your broker. Be sure to capture the broker host name, for example:


You should also modify the Security Group for the broker by clicking on its Security Group ID. Click the Edit button and then click Add Rule to allow inbound traffic on port 8162 for your IP address.

Deploying and scheduling the Lambda function

To simplify the deployment of this example, I’ve provided an AWS Serverless Application Model (SAM) template that deploys the sample function and DynamoDB table, and schedules the function to be invoked every five minutes. Detailed instructions can be found with sample code on GitHub in the amazonmq-invoke-aws-lambda repository, with sample code. I discuss a few key aspects in this post.

First, SAM makes it easy to deploy and schedule invocation of our function:

	Type: AWS::Serverless::Function
		CodeUri: subscriber/
		Handler: index.handler
		Runtime: nodejs6.10
		Role: !GetAtt SubscriberFunctionRole.Arn
		Timeout: 15
				HOST: !Ref AmazonMQHost
				LOGIN: !Ref AmazonMQLogin
				PASSWORD: !Ref AmazonMQPassword
				QUEUE_NAME: !Ref AmazonMQQueueName
				WORKER_FUNCTIOn: !Ref WorkerFunction
				Type: Schedule
					Schedule: rate(5 minutes)

Type: AWS::Serverless::Function
		CodeUri: worker/
		Handler: index.handler
		Runtime: nodejs6.10
Role: !GetAtt WorkerFunctionRole.Arn
				TABLE_NAME: !Ref MessagesTable

In the code, you include the URI, user name, and password for your newly created Amazon MQ broker. These allow the function to poll the broker for new messages on the sample queue.

The sample Lambda function is written in Node.js, but clients exist for a number of programming languages.

stomp.connect(options, (error, client) => {
	if (error) { /* do something */ }

	let headers = {
		destination: ‘/queue/SAMPLE_QUEUE’,
		ack: ‘auto’

	client.subscribe(headers, (error, message) => {
		if (error) { /* do something */ }

		message.readString(‘utf-8’, (error, body) => {
			if (error) { /* do something */ }

			let params = {
				FunctionName: MyWorkerFunction,
				Payload: JSON.stringify({
					message: body,
					timestamp: Date.now()

			let lambda = new AWS.Lambda()
			lambda.invoke(params, (error, data) => {
				if (error) { /* do something */ }

Sending a sample message

For the purpose of this example, use the Amazon MQ console to send a test message. Navigate to the details page for your broker.

About midway down the page, choose ActiveMQ Web Console. Next, choose Manage ActiveMQ Broker to launch the admin console. When you are prompted for a user name and password, use the credentials created earlier.

At the top of the page, choose Send. From here, you can send a sample message from the broker to subscribers. For this example, this is how you generate traffic to test the end-to-end system. Be sure to set the Destination value to “SAMPLE_QUEUE.” The message body can contain any text. Choose Send.

You now have a Lambda function polling for messages on the broker. To verify that your function is working, you can confirm in the DynamoDB console that the message was successfully received and processed by the sample Lambda function.

First, choose Tables on the left and select the table name “amazonmq-messages” in the middle section. With the table detail in view, choose Items. If the function was successful, you’ll find a new entry similar to the following:

If there is no message in DynamoDB, check again in a few minutes or review the CloudWatch Logs group for Lambda functions that contain debug messages.

Alternative approaches

Beyond the approach described here, you may consider other approaches as well. For example, you could use an intermediary system such as Apache Flume to pass messages from the broker to Lambda or deploy Apache Camel to trigger Lambda via a POST to API Gateway. There are trade-offs to each of these approaches. My goal in using CloudWatch Events was to introduce an easily repeatable pattern familiar to many Lambda developers.


I hope that you have found this example of how to integrate AWS Lambda with Amazon MQ useful. If you have expertise or legacy systems that leverage APIs such as JMS, you may find this useful as you incorporate serverless concepts in your enterprise architectures.

To learn more, see the Amazon MQ website and Developer Guide. You can try Amazon MQ for free with the AWS Free Tier, which includes up to 750 hours of a single-instance mq.t2.micro broker and up to 1 GB of storage per month for one year.

Implementing Dynamic ETL Pipelines Using AWS Step Functions

Post Syndicated from Tara Van Unen original https://aws.amazon.com/blogs/compute/implementing-dynamic-etl-pipelines-using-aws-step-functions/

This post contributed by:
Wangechi Dole, AWS Solutions Architect
Milan Krasnansky, ING, Digital Solutions Developer, SGK
Rian Mookencherry, Director – Product Innovation, SGK

Data processing and transformation is a common use case you see in our customer case studies and success stories. Often, customers deal with complex data from a variety of sources that needs to be transformed and customized through a series of steps to make it useful to different systems and stakeholders. This can be difficult due to the ever-increasing volume, velocity, and variety of data. Today, data management challenges cannot be solved with traditional databases.

Workflow automation helps you build solutions that are repeatable, scalable, and reliable. You can use AWS Step Functions for this. A great example is how SGK used Step Functions to automate the ETL processes for their client. With Step Functions, SGK has been able to automate changes within the data management system, substantially reducing the time required for data processing.

In this post, SGK shares the details of how they used Step Functions to build a robust data processing system based on highly configurable business transformation rules for ETL processes.

SGK: Building dynamic ETL pipelines

SGK is a subsidiary of Matthews International Corporation, a diversified organization focusing on brand solutions and industrial technologies. SGK’s Global Content Creation Studio network creates compelling content and solutions that connect brands and products to consumers through multiple assets including photography, video, and copywriting.

We were recently contracted to build a sophisticated and scalable data management system for one of our clients. We chose to build the solution on AWS to leverage advanced, managed services that help to improve the speed and agility of development.

The data management system served two main functions:

  1. Ingesting a large amount of complex data to facilitate both reporting and product funding decisions for the client’s global marketing and supply chain organizations.
  2. Processing the data through normalization and applying complex algorithms and data transformations. The system goal was to provide information in the relevant context—such as strategic marketing, supply chain, product planning, etc. —to the end consumer through automated data feeds or updates to existing ETL systems.

We were faced with several challenges:

  • Output data that needed to be refreshed at least twice a day to provide fresh datasets to both local and global markets. That constant data refresh posed several challenges, especially around data management and replication across multiple databases.
  • The complexity of reporting business rules that needed to be updated on a constant basis.
  • Data that could not be processed as contiguous blocks of typical time-series data. The measurement of the data was done across seasons (that is, combination of dates), which often resulted with up to three overlapping seasons at any given time.
  • Input data that came from 10+ different data sources. Each data source ranged from 1–20K rows with as many as 85 columns per input source.

These challenges meant that our small Dev team heavily invested time in frequent configuration changes to the system and data integrity verification to make sure that everything was operating properly. Maintaining this system proved to be a daunting task and that’s when we turned to Step Functions—along with other AWS services—to automate our ETL processes.

Solution overview

Our solution included the following AWS services:

  • AWS Step Functions: Before Step Functions was available, we were using multiple Lambda functions for this use case and running into memory limit issues. With Step Functions, we can execute steps in parallel simultaneously, in a cost-efficient manner, without running into memory limitations.
  • AWS Lambda: The Step Functions state machine uses Lambda functions to implement the Task states. Our Lambda functions are implemented in Java 8.
  • Amazon DynamoDB provides us with an easy and flexible way to manage business rules. We specify our rules as Keys. These are key-value pairs stored in a DynamoDB table.
  • Amazon RDS: Our ETL pipelines consume source data from our RDS MySQL database.
  • Amazon Redshift: We use Amazon Redshift for reporting purposes because it integrates with our BI tools. Currently we are using Tableau for reporting which integrates well with Amazon Redshift.
  • Amazon S3: We store our raw input files and intermediate results in S3 buckets.
  • Amazon CloudWatch Events: Our users expect results at a specific time. We use CloudWatch Events to trigger Step Functions on an automated schedule.

Solution architecture

This solution uses a declarative approach to defining business transformation rules that are applied by the underlying Step Functions state machine as data moves from RDS to Amazon Redshift. An S3 bucket is used to store intermediate results. A CloudWatch Event rule triggers the Step Functions state machine on a schedule. The following diagram illustrates our architecture:

Here are more details for the above diagram:

  1. A rule in CloudWatch Events triggers the state machine execution on an automated schedule.
  2. The state machine invokes the first Lambda function.
  3. The Lambda function deletes all existing records in Amazon Redshift. Depending on the dataset, the Lambda function can create a new table in Amazon Redshift to hold the data.
  4. The same Lambda function then retrieves Keys from a DynamoDB table. Keys represent specific marketing campaigns or seasons and map to specific records in RDS.
  5. The state machine executes the second Lambda function using the Keys from DynamoDB.
  6. The second Lambda function retrieves the referenced dataset from RDS. The records retrieved represent the entire dataset needed for a specific marketing campaign.
  7. The second Lambda function executes in parallel for each Key retrieved from DynamoDB and stores the output in CSV format temporarily in S3.
  8. Finally, the Lambda function uploads the data into Amazon Redshift.

To understand the above data processing workflow, take a closer look at the Step Functions state machine for this example.

We walk you through the state machine in more detail in the following sections.


To get started, you need to:

  • Create a schedule in CloudWatch Events
  • Specify conditions for RDS data extracts
  • Create Amazon Redshift input files
  • Load data into Amazon Redshift

Step 1: Create a schedule in CloudWatch Events
Create rules in CloudWatch Events to trigger the Step Functions state machine on an automated schedule. The following is an example cron expression to automate your schedule:

In this example, the cron expression invokes the Step Functions state machine at 3:00am and 2:00pm (UTC) every day.

Step 2: Specify conditions for RDS data extracts
We use DynamoDB to store Keys that determine which rows of data to extract from our RDS MySQL database. An example Key is MCS2017, which stands for, Marketing Campaign Spring 2017. Each campaign has a specific start and end date and the corresponding dataset is stored in RDS MySQL. A record in RDS contains about 600 columns, and each Key can represent up to 20K records.

A given day can have multiple campaigns with different start and end dates running simultaneously. In the following example DynamoDB item, three campaigns are specified for the given date.

The state machine example shown above uses Keys 31, 32, and 33 in the first ChoiceState and Keys 21 and 22 in the second ChoiceState. These keys represent marketing campaigns for a given day. For example, on Monday, there are only two campaigns requested. The ChoiceState with Keys 21 and 22 is executed. If three campaigns are requested on Tuesday, for example, then ChoiceState with Keys 31, 32, and 33 is executed. MCS2017 can be represented by Key 21 and Key 33 on Monday and Tuesday, respectively. This approach gives us the flexibility to add or remove campaigns dynamically.

Step 3: Create Amazon Redshift input files
When the state machine begins execution, the first Lambda function is invoked as the resource for FirstState, represented in the Step Functions state machine as follows:

"Comment": ” AWS Amazon States Language.", 
  "StartAt": "FirstState",
"States": { 
  "FirstState": {
"Type": "Task",
"Resource": "arn:aws:lambda:xx-xxxx-x:XXXXXXXXXXXX:function:Start",
    "Next": "ChoiceState" 

As described in the solution architecture, the purpose of this Lambda function is to delete existing data in Amazon Redshift and retrieve keys from DynamoDB. In our use case, we found that deleting existing records was more efficient and less time-consuming than finding the delta and updating existing records. On average, an Amazon Redshift table can contain about 36 million cells, which translates to roughly 65K records. The following is the code snippet for the first Lambda function in Java 8:

public class LambdaFunctionHandler implements RequestHandler<Map<String,Object>,Map<String,String>> {
    Map<String,String> keys= new HashMap<>();
    public Map<String, String> handleRequest(Map<String, Object> input, Context context){
       Properties config = getConfig(); 
       // 1. Cleaning Redshift Database
       new RedshiftDataService(config).cleaningTable(); 
       // 2. Reading data from Dynamodb
       List<String> keyList = new DynamoDBDataService(config).getCurrentKeys();
       for(int i = 0; i < keyList.size(); i++) {
           keys.put(”key" + (i+1), keyList.get(i)); 
       keys.put(”key" + T,String.valueOf(keyList.size()));
       // 3. Returning the key values and the key count from the “for” loop
       return (keys);

The following JSON represents ChoiceState.

"ChoiceState": {
   "Type" : "Choice",
   "Choices": [ 

      "Variable": "$.keyT",
     "StringEquals": "3",
     "Next": "CurrentThreeKeys" 

     "Variable": "$.keyT",
    "StringEquals": "2",
    "Next": "CurrentTwooKeys" 
 "Default": "DefaultState"

The variable $.keyT represents the number of keys retrieved from DynamoDB. This variable determines which of the parallel branches should be executed. At the time of publication, Step Functions does not support dynamic parallel state. Therefore, choices under ChoiceState are manually created and assigned hardcoded StringEquals values. These values represent the number of parallel executions for the second Lambda function.

For example, if $.keyT equals 3, the second Lambda function is executed three times in parallel with keys, $key1, $key2 and $key3 retrieved from DynamoDB. Similarly, if $.keyT equals two, the second Lambda function is executed twice in parallel.  The following JSON represents this parallel execution:

"CurrentThreeKeys": { 
  "Type": "Parallel",
  "Next": "NextState",
  "Branches": [ 

     "StartAt": “key31",
    "States": { 
       “key31": {

          "Type": "Task",
        "InputPath": "$.key1",
        "Resource": "arn:aws:lambda:xx-xxxx-x:XXXXXXXXXXXX:function:Execution",
        "End": true 

     "StartAt": “key32",
    "States": { 
     “key32": {

        "Type": "Task",
       "InputPath": "$.key2",
         "Resource": "arn:aws:lambda:xx-xxxx-x:XXXXXXXXXXXX:function:Execution",
       "End": true 

      "StartAt": “key33",
       "States": { 
          “key33": {

                "Type": "Task",
             "InputPath": "$.key3",
             "Resource": "arn:aws:lambda:xx-xxxx-x:XXXXXXXXXXXX:function:Execution",
           "End": true 

Step 4: Load data into Amazon Redshift
The second Lambda function in the state machine extracts records from RDS associated with keys retrieved for DynamoDB. It processes the data then loads into an Amazon Redshift table. The following is code snippet for the second Lambda function in Java 8.

public class LambdaFunctionHandler implements RequestHandler<String, String> {
 public static String key = null;

public String handleRequest(String input, Context context) { 
   //1. Getting basic configurations for the next classes + s3 client Properties
   config = getConfig();

   AmazonS3 s3 = AmazonS3ClientBuilder.defaultClient(); 
   // 2. Export query results from RDS into S3 bucket 
   new RdsDataService(config).exportDataToS3(s3,key); 
   // 3. Import query results from S3 bucket into Redshift 
    new RedshiftDataService(config).importDataFromS3(s3,key); 
   return "SUCCESS"; 

After the data is loaded into Amazon Redshift, end users can visualize it using their preferred business intelligence tools.

Lessons learned

  • At the time of publication, the 1.5–GB memory hard limit for Lambda functions was inadequate for processing our complex workload. Step Functions gave us the flexibility to chunk our large datasets and process them in parallel, saving on costs and time.
  • In our previous implementation, we assigned each key a dedicated Lambda function along with CloudWatch rules for schedule automation. This approach proved to be inefficient and quickly became an operational burden. Previously, we processed each key sequentially, with each key adding about five minutes to the overall processing time. For example, processing three keys meant that the total processing time was three times longer. With Step Functions, the entire state machine executes in about five minutes.
  • Using DynamoDB with Step Functions gave us the flexibility to manage keys efficiently. In our previous implementations, keys were hardcoded in Lambda functions, which became difficult to manage due to frequent updates. DynamoDB is a great way to store dynamic data that changes frequently, and it works perfectly with our serverless architectures.


With Step Functions, we were able to fully automate the frequent configuration updates to our dataset resulting in significant cost savings, reduced risk to data errors due to system downtime, and more time for us to focus on new product development rather than support related issues. We hope that you have found the information useful and that it can serve as a jump-start to building your own ETL processes on AWS with managed AWS services.

For more information about how Step Functions makes it easy to coordinate the components of distributed applications and microservices in any workflow, see the use case examples and then build your first state machine in under five minutes in the Step Functions console.

If you have questions or suggestions, please comment below.