Tag Archives: Amazon CloudWatch

Best practices working with self-hosted GitHub Action runners at scale on AWS

Post Syndicated from Shilpa Sharma original https://aws.amazon.com/blogs/devops/best-practices-working-with-self-hosted-github-action-runners-at-scale-on-aws/

Overview

GitHub Actions is a continuous integration and continuous deployment platform that enables the automation of build, test and deployment activities for your workload. GitHub Self-Hosted Runners provide a flexible and customizable option to run your GitHub Action pipelines. These runners allow you to run your builds on your own infrastructure, giving you control over the environment in which your code is built, tested, and deployed. This reduces your security risk and costs, and gives you the ability to use specific tools and technologies that may not be available in GitHub hosted runners. In this blog, I explore security, performance and cost best practices to take into consideration when deploying GitHub Self-Hosted Runners to your AWS environment.

Best Practices

Understand your security responsibilities

GitHub Self-hosted runners, by design, execute code defined within a GitHub repository, either through the workflow scripts or through the repository build process. You must understand that the security of your AWS runner execution environments are dependent upon the security of your GitHub implementation. Whilst a complete overview of GitHub security is outside the scope of this blog, I recommended that before you begin integrating your GitHub environment with your AWS environment, you review and understand at least the following GitHub security configurations.

  • Federate your GitHub users, and manage the lifecycle of identities through a directory.
  • Limit administrative privileges of GitHub repositories, and restrict who is able to administer permissions, write to repositories, modify repository configurations or install GitHub Apps.
  • Limit control over GitHub Actions runner registration and group settings
  • Limit control over GitHub workflows, and follow GitHub’s recommendations on using third-party actions
  • Do not allow public repositories access to self-hosted runners

Reduce security risk with short-lived AWS credentials

Make use of short-lived credentials wherever you can. They expire by default within 1 hour, and you do not need to rotate or explicitly revoke them. Short lived credentials are created by the AWS Security Token Service (STS). If you use federation to access your AWS account, assume roles, or use Amazon EC2 instance profiles and Amazon ECS task roles, you are using STS already!

In almost all cases, you do not need long-lived AWS Identity and Access Management (IAM) credentials (access keys) even for services that do not “run” on AWS – you can extend IAM roles to workloads outside of AWS without requiring you to manage long-term credentials. With GitHub Actions, we suggest you use OpenID Connect (OIDC). OIDC is a decentralized authentication protocol that is natively supported by STS using sts:AssumeRoleWithWebIdentity, GitHub and many other providers. With OIDC, you can create least-privilege IAM roles tied to individual GitHub repositories and their respective actions. GitHub Actions exposes an OIDC provider to each action run that you can utilize for this purpose.

Short lived AWS credentials with GitHub self-hosted runners

Short lived AWS credentials with GitHub self-hosted runners

If you have many repositories that you wish to grant an individual role to, you may run into a hard limit of the number of IAM roles in a single account. While I advocate solving this problem with a multi-account strategy, you can alternatively scale this approach by:

  • using attribute based access control (ABAC) to match claims in the GitHub token (such as repository name, branch, or team) to the AWS resource tags.
  • using role based access control (RBAC) by logically grouping the repositories in GitHub into Teams or applications to create fewer subset of roles.
  • use an identity broker pattern to vend credentials dynamically based on the identity provided to the GitHub workflow.

Use Ephemeral Runners

Configure your GitHub Action runners to run in “ephemeral” mode, which creates (and destroys) individual short-lived compute environments per job on demand. The short environment lifespan and per-build isolation reduces the risk of data leakage , even in multi-tenanted continuous integration environments, as each build job remains isolated from others on the underlying host.

As each job runs on a new environment created on demand, there is no need for a job to wait for an idle runner, simplifying auto-scaling. With the ability to scale runners on demand, you do not need to worry about turning build infrastructure off when it is not needed (for example out of office hours), giving you a cost-efficient setup. To optimize the setup further, consider allowing developers to tag workflows with instance type tags and launch specific instance types that are optimal for respective workflows.

There are a few considerations to take into account when using ephemeral runners:

  • A job will remain queued until the runner EC2 instance has launched and is ready. This can take up to 2 minutes to complete. To speed up this process, consider using an optimized AMI with all prerequisites installed.
  • Since each job is launched on a fresh runner, utilizing caching on the runner is not possible. For example, Docker images and libraries will always be pulled from source.

Use Runner Groups to isolate your runners based on security requirements

By using ephemeral runners in a single GitHub runner group, you are creating a pool of resources in the same AWS account that are used by all repositories sharing this runner group. Your organizational security requirements may dictate that your execution environments must be isolated further, such as by repository or by environment (such as dev, test, prod).

Runner groups allow you to define the runners that will execute your workflows on a repository-by-repository basis. Creating multiple runner groups not only allow you to provide different types of compute environments, but allow you to place your workflow executions in locations within AWS that are isolated from each other. For example, you may choose to locate your development workflows in one runner group and test workflows in another, with each ephemeral runner group being deployed to a different AWS account.

Runners by definition execute code on behalf of your GitHub users. At a minimum, I recommend that your ephemeral runner groups are contained within their own AWS account and that this AWS account has minimal access to other organizational resources. When access to organizational resources is required, this can be given on a repository-by-repository basis through IAM role assumption with OIDC, and these roles can be given least-privilege access to the resources they require.

Optimize runner start up time using Amazon EC2 warm-pools

Ephemeral runners provide strong build isolation, simplicity and security. Since the runners are launched on demand, the job will be required to wait for the runner to launch and register itself with GitHub. While this usually happens in under 2 minutes, this wait time might not be acceptable in some scenarios.

We can use a warm pool of pre-registered ephemeral runners to reduce the wait time. These runners will listen to the incoming GitHub workflow events actively and as soon as an incoming workflow event is queued, it is picked up readily by the warm pool of registered EC2 runners.

While there can be multiple strategies to manage the warm pool, I recommend the following strategy which uses AWS Lambda for scaling up and scaling down the ephemeral runners:

GitHub self-hosted runners warm pool flow

GitHub self-hosted runners warm pool flow

A GitHub workflow event is created on a trigger like push of code in a master repository or a merge of pull request. This event triggers a Lambda function via webhook and Amazon API Gateway endpoint. The Lambda function helps in validating the GitHub workflow event payload and log events for observability & building metrics. It can be used optionally to replenish the warm pool. There are separate backend Lambda functions to launch, scale up and scale down the warm pool of EC2 instances. The EC2 instances or runners are registered with GitHub at the time of launch. The registered runners listens for incoming GitHub work flow events using GitHub’s internal job queue and as soon as workflow events are triggered, its assigned by GitHub to one of the runners in warm pool for job execution. The runner is automatically de-registered once the job completes. A job can be a build, or deploy request as defined in your GitHub workflow.

With warm pool in place, it is expected to help reduce wait time by 70-80%.

Considerations

  • Increased complexity as there is a possibility of over provisioning runners. This will depend on how long a runner EC2 instance requires to launch and reach a ready state and how frequently the scale up Lambda is configured to run. For example, if the scale up Lambda runs every 1 minute and the EC2 runner requires 2 minutes to launch, then the scale up Lambda will launch 2 instances. The mitigation is to use Auto scaling groups to manage the EC2 warm pool and desired capacity with predictive scaling policies tying back to incoming GitHub workflow events i.e. build job requests.
  • This strategy may have to be revised when supporting Windows or Mac based runners given the spin up times can vary.

Use an optimized AMI to speed up the launch of GitHub self-hosted runners

Amazon Machine Images (AMI) provide a pre-configured, optimized image that can be used to launch the runner EC2 instance. By using AMIs, you will be able to reduce the launch time of a new runner since dependencies and tools are already installed. Consistency across builds is guaranteed due to all instances running the same version of dependencies and tools. Runners will benefit from increased stability and security compliance as images are tested and approved before being distributed for use as runner instances.

When building an AMI for use as a GitHub self-hosted runner the following considerations need to be made:

  • Choosing the right OS base image for the builds. This will depend on your tech stack and toolset.
  • Install the GitHub runner app as part of the image. Ensure automatic runner updates are enabled to reduce the overhead of managing running versions. In case a specific runner version must be used you can disable automatic runner updates to avoid untested changes. Keep in mind, if disabled, a runner will need to be updated manually within 30 days of a new version becoming available.
  • Install build tools and dependencies from trusted sources.
  • Ensure runner logs are captured and forwarded to your security information and event management (SIEM) of choice.
  • The runner requires internet connectivity to access GitHub. This may require configuring proxy settings on the instance depending on your networking setup.
  • Configure any artifact repositories the runner requires. This includes sources and authentication.
  • Automate the creation of the AMI using tools such as EC2 Image Builder to achieve consistency.

Use Spot instances to save costs

The cost associated with scaling up the runners as well as maintaining a hot pool can be minimized using Spot Instances, which can result in savings up to 90% compared to On-Demand prices. However, there could be requirements where we can have longer running builds or batch jobs that cannot tolerate the spot instance terminating on 2 minutes notice. So, having a mixed pool of instances will be a good option where such jobs should be routed to on-demand EC2 instances and the rest on the Spot instances to cater for diverse build needs. This can be done by assigning labels to the runner during launch /registration. In that case, the on-demand instances will be launched and we can a savings plan in place to get cost benefits.

Record runner metrics using Amazon CloudWatch for Observability

It is vital for the observability of the overall platform to generate metrics for the EC2 based GitHub self-hosted runners. Examples of the GitHub runners metrics can be: the number of GitHub workflow events queued or completed in a minute, or number of EC2 runners up and available in the warm pool etc.

We can log the triggered workflow events and runner logs in Amazon CloudWatch and then use CloudWatch embedded metrics to collect metrics such as number of workflow events queued, in progress and completed. Using elements like “started_at” and “completed_at” timings which are part of workflow event payload we can calculate build wait time.

As an example, below is the sample incoming GitHub workflow event logged in Amazon Cloud Watch Logs

<p> </p><p><code>{</code></p><p><code>"hostname": "xxx.xxx.xxx.xxx",</code></p><p><code>"requestId": "aafddsd55-fgcf555",</code></p><p><code>"date": "2022-10-11T05:50:35.816Z",</code></p><p><code>"logLevel": "info",</code></p><p><code>"logLevelId": 3,</code></p><p><code>"filePath": "index.js",</code></p><p><code>"fullFilePath": "/var/task/index.js",</code></p><p><code>"fileNa<a class="ab-item" href="https://aws-blogs-prod.amazon.com/devops/" aria-haspopup="true">AWS DevOps Blog</a>me": "index.js",</code></p><p><code>"lineNumber": 83889,</code></p><p><code>"columnNumber": 12,</code></p><p><code>"isConstructor": false,</code></p><p><code>"functionName": "handle",</code></p><p><code>"argumentsArray": [</code></p><p><code>"Processing Github event",</code></p><p><code>"{\"event\":\"workflow_job\",\"repository\":\"testorg-poc/github-actions-test-repo\",\"action\":\"queued\",\"name\":\"jobname-buildanddeploy\",\"status\":\"queued\",\"started_at\":\"2022-10-11T05:50:33Z\",\"completed_at\":null,\"conclusion\":null}"</code></p><p><code>]</code></p><p><code>}</code></p>

In order to use the logged elements of above log into metrics by capturing \”status\”:\”queued\”,\”repository\”:\”testorg-poc/github-actions-test-repo\c, \”name\”:\”jobname-buildanddeploy\” ,and workflow \”event\” , one can build embedded metrics in the application code or AWS metrics Lambda using any of the cloud watch metrics client library Creating logs in embedded metric format using the client libraries – Amazon CloudWatch based on the language of your choice listed.c

Essentially what one of those libraries will do under the hood is map elements from Log event into dimension fields so cloud watch can then read and generate a metric using that.

console.log(<br />      JSON.stringify({<br />        message: '[Embedded Metric]', // Identifier for metric logs in CW logs<br />        build_event_metric: 1, // Metric Name and value<br />        status: `${status}`, // Dimension name and value<br />        eventName: `${eventName}`,<br />        repository: `${repository}`,<br />        name: `${name}`,<br />        <br />        _aws: {<br />          Timestamp: Date.now(),<br />          CloudWatchMetrics: [<br />            {<br />              Namespace: `demo_2`,<br />              Dimensions: [['status','eventName','repository','name']],<br />              Metrics: [<br />                {<br />                  Name: 'build_event_metric',<br />                  Unit: 'Count',<br />                },<br />              ],<br />            },<br />          ],<br />        },<br />      })<br />    );

A sample architecture:

Consumption of GitHub webhook events

Consumption of GitHub webhook events

The cloud watch metrics can be published to your dashboards or forwarded to any external tool based on requirements. Once we have metrics, CloudWatch alarms and notifications can be configured to manage pool exhaustion.

Conclusion

In this blog post, we outlined several best practices covering security, scalability and cost efficiency when using GitHub Actions with EC2 self-hosted runners on AWS. We covered how using short-lived credentials combined with ephemeral runners will reduce security and build contamination risks. We also showed how runners can be optimized for faster startup and job execution AMIs and warm EC2 pools. Last but not least, cost efficiencies can be maximized by using Spot instances for runners in the right scenarios.

Resources:

AWS Weekly Roundup: Amazon EC2 U7i Instances, Bedrock Converse API, AWS World IPv6 Day and more (June 3, 2024)

Post Syndicated from Channy Yun original https://aws.amazon.com/blogs/aws/aws-weekly-roundup-amazon-ec2-u7i-instances-bedrock-converse-api-aws-world-ipv6-day-and-more-june-3-2024/

Life is not always happy, there are difficult times. However, we can share our joys and sufferings with those we work with. The AWS Community is no exception.

Jeff Barr introduced two members of the AWS community who are dealing with health issues. Farouq Mousa is an AWS Community Builder and fighting brain cancer. Allen Helton is an AWS Serverless Hero and his young daughter is fighting leukemia.

Please donate to support Farauq and Olivia, Allen’s daughter to overcome their disease.

Last week’s launches
Here are some launches that got my attention:

Amazon EC2 high memory U7i Instances – These instances with up to 32 TiB of DDR5 memory and 896 vCPUs are powered by custom fourth generation Intel Xeon Scalable Processors (Sapphire Rapids). These high memory instances are designed to support large, in-memory databases including SAP HANA, Oracle, and SQL Server. To learn more, visit Jeff’s blog post.

New Amazon Connect analytics data lake – You can use a single source for contact center data including contact records, agent performance, Contact Lens insights, and more — eliminating the need to build and maintain complex data pipelines. Your organization can create your own custom reports using Amazon Connect data or combine data queried from third-party sources. To learn more, visit Donnie’s blog post.

Amazon Bedrock Converse API – This API provides developers a consistent way to invoke Amazon Bedrock models removing the complexity to adjust for model-specific differences such as inference parameters. With this API, you can write a code once and use it seamlessly with different models in Amazon Bedrock. To learn more, visit Dennis’s blog post to get started.

New Document widget for PartyRock – You can build, use, and share generative AI-powered apps for fun and for boosting personal productivity, using PartyRock. Its widgets display content, accept input, connect with other widgets, and generate outputs like text, images, and chats using foundation models. You can now use new document widget to integrate text content from files and documents directly into a PartyRock app.

30 days of alarm history in Amazon CloudWatch – You can view the history of your alarm state changes for up to 30 days prior. Previously, CloudWatch provided 2 weeks of alarm history. This extended history makes it easier to observe past behavior and review incidents over a longer period of time. To learn more, visit the CloudWatch alarms documentation section.

10x faster startup time in Amazon SageMaker Canvas – You can launch SageMaker Canvas in less than a minute and get started with your visual, no-code interface for machine learning 10x faster than before. Now, all new user profiles created in existing or new SageMaker domains can experience this accelerated startup time.

For a full list of AWS announcements, be sure to keep an eye on the What’s New at AWS page.

Other AWS news
Here are some additional news items and a Twitch show that you might find interesting:

Let us manage your relational database! – Jeff Barr ran a poll to better understand why some AWS customers still choose to host their own databases in the cloud. Working backwards, he highlights four issues that AWS managed database services address. Consider these before hosting your own database.

Amazon Bedrock Serverless Prompt Chaining – This repository provides examples of using AWS Step Functions and Amazon Bedrock to build complex, serverless, and highly scalable generative AI applications with prompt chaining.

AWS Merch Store Spring Sale – Do you want to buy AWS branded t-shirts, hats, bags, and so on? Get 15% off on all items now through June 7th.

Upcoming AWS events
Check your calendars and sign up for these AWS events:

AWS World IPv6 Day — Join us a free in-person celebration event on June 6, for technical presentations from AWS experts plus a workshop and whiteboarding session. You will learn how to get started with IPv6 and hear from customers who have started on the journey of IPv6 adoption. Check out your near city: San Francisco, Seattle, New YorkLondon, Mumbai, Bangkok, Singapore, Kuala Lumpur, Beijing, Manila, and Sydney.

AWS Summits — Join free online and in-person events that bring the cloud computing community together to connect, collaborate, and learn about AWS. Register in your nearest city: Stockholm (June 4), Madrid (June 5), and Washington, DC (June 26–27).

AWS re:Inforce — Join us for AWS re:Inforce (June 10–12) in Philadelphia, PA. AWS re:Inforce is a learning conference focused on AWS security solutions, cloud security, compliance, and identity. Connect with the AWS teams that build the security tools and meet AWS customers to learn about their security journeys.

AWS Community Days — Join community-led conferences that feature technical discussions, workshops, and hands-on labs led by expert AWS users and industry leaders from around the world: Midwest | Columbus (June 13), Sri Lanka (June 27), Cameroon (July 13), New Zealand (August 15), Nigeria (August 24), and New York (August 28).

You can browse all upcoming in-person and virtual events.

That’s all for this week. Check back next Monday for another Weekly Roundup!

Channy

This post is part of our Weekly Roundup series. Check back each week for a quick roundup of interesting news and announcements from AWS!

Amazon CloudWatch Internet Weather Map – View and analyze internet health

Post Syndicated from Jeff Barr original https://aws.amazon.com/blogs/aws/amazon-cloudwatch-internet-weather-map-view-and-analyze-internet-health/

The Internet has a plethora of moving parts: routers, switches, hubs, terrestrial and submarine cables, and connectors on the hardware side, and complex protocol stacks and configurations on the software side. When something goes wrong that slows or disrupts the Internet in a way that affects your customers, you want to be able to localize and understand the issue as quickly as possible.

New Map
The new Amazon CloudWatch Internet Weather Map is here to help! Built atop of collection of global monitors operated by AWS, you get a broad, global view of Internet weather, with the ability to zoom in and understand performance and availability issues that affect a particular city. To access the map, open the CloudWatch Console, expand Network monitoring on the left, and click Internet Monitor. The map appears and displays weather for the entire world:

The red and yellow circles indicate current, active issues that affect availability or performance, respectively. The grey circles represent issues that have been resolved within the last 24 hours, and the blue diamonds represent AWS regions. The map will automatically refresh every 15 minutes if you leave it on the screen.

Each issue affects a specific city-network, representing a combination of a location where clients access AWS resources, and the Autonomous System Number (ASN) that was used to access the resources. ASNs typically represent individual Internet Service Providers (ISPs).

The list to the right of the map shows active events at the top, followed by events that have been resolved in the recent past, looking back up to 24 hours:

I can hover my mouse over any of the indicators to see the list of city-networks in the geographic area:

If I zoom in a step or two, I can see that those city-networks are spread out over the United States:

I can zoom in even further and see a single city-network:

This information is also available programmatically. The new ListInternetEvents function returns up to 100 performance or availability events per call, with optional filtering by time range, status (ACTIVE or RESOLVED), or type (PERFORMANCE or AVAILABILITY). Each event includes full details including latitude and longitude.

The new map is accessible from all AWS regions and there is no charge to use it. Going forward, we have a lot of powerful additions on the roadmap, subject to prioritization based on your feedback. Right now we are thinking about:

  1. Displaying causes of certain types of outages such as DDoS attacks, BGP route leaks, and issues with route interconnects.
  2. Adding a view that is specific to a chosen ISP.
  3. Displaying the impact to public SaaS applications.

Please feel free to send feedback on this feature to [email protected] .

CloudWatch Internet Monitor
The information in the map applies to everyone who makes use of applications built on AWS. If you want to understand how internet weather affects your particular AWS applications and to take advantage of other features such as health event notification and traffic insights, you can make use of CloudWatch Internet Monitor. As my colleague Sébastien wrote when he launched this feature in late 2022:

You told us one of your challenges when monitoring internet-facing applications is to gather data outside of AWS to build a realistic picture of how your application behaves for your customers connected to multiple and geographically distant internet providers. Capturing and monitoring data about internet traffic before it reaches your infrastructure is either difficult or very expensive.

After you review the map, you can click Create monitor to get started with CloudWatch Internet Monitor:

After that you enter a name for your monitor, choose the AWS resources (VPCs, CloudFront distributions, Network Load Balancers, and Amazon WorkSpace Directories) to monitor, then select the desired percentage of internet-facing traffic to monitor. The monitor will begin to operate within minutes, using entries from your VPC Flow Logs, CloudFront Access Logs, and other telemetry to identify the most relevant city-networks.

Here are some resources to help you learn more about this feature:

Jeff;

Serverless ICYMI Q1 2024

Post Syndicated from Julian Wood original https://aws.amazon.com/blogs/compute/serverless-icymi-q1-2024/

Welcome to the 25th edition of the AWS Serverless ICYMI (in case you missed it) quarterly recap. Every quarter, we share all the most recent product launches, feature enhancements, blog posts, webinars, live streams, and other interesting things that you might have missed!

In case you missed our last ICYMI, check out what happened last quarter here.

2024 Q1 calendar

2024 Q1 calendar

Adobe Summit

At the Adobe Summit, the AWS Serverless Developer Advocacy team showcased a solution developed for the NFL using AWS serverless technologies and Adobe Photoshop APIs. The system automates image processing tasks, including background removal and dynamic resizing, by integrating AWS Step Functions, AWS Lambda, Amazon EventBridge, and AI/ML capabilities via Amazon Rekognition. This solution reduced image processing time from weeks to minutes and saved the NFL significant costs. Combining cloud-based serverless architectures with advanced machine learning and API technologies can optimize digital workflows for cost-effective and agile digital asset management.

Adobe Summit ServerlessVideo

Adobe Summit ServerlessVideo

ServerlessVideo is a demo application to stream live videos and also perform advanced post-video processing. It uses several AWS services, including Step Functions, Lambda, EventBridge, Amazon ECS, and Amazon Bedrock in a serverless architecture that makes it fast, flexible, and cost-effective. The team used ServerlessVideo to interview attendees about the conference experience and Adobe and partners about how they use Adobe. Learn more about the project and watch videos from Adobe Summit 2024 at video.serverlessland.com.

AWS Lambda

AWS launched support for the latest long-term support release of .NET 8, which includes API enhancements, improved Native Ahead of Time (Native AOT) support, and improved performance.

AWS Lambda .NET 8

AWS Lambda .NET 8

Learn how to compare design approaches for building serverless microservices. This post covers the trade-offs to consider with various application architectures. See how you can apply single responsibility, Lambda-lith, and read and write functions.

The AWS Serverless Java Container has been updated. This makes it easier to modernize a legacy Java application written with frameworks such as Spring, Spring Boot, or JAX-RS/Jersey in Lambda with minimal code changes.

AWS Serverless Java Container

AWS Serverless Java Container

Lambda has improved the responsiveness for configuring Event Source Mappings (ESMs) and Amazon EventBridge Pipes with event sources such as self-managed Apache Kafka, Amazon Managed Streaming for Apache Kafka (MSK), Amazon DocumentDB, and Amazon MQ.

Chaos engineering is a popular practice for building confidence in system resilience. However, many existing tools assume the ability to alter infrastructure configurations, and cannot be easily applied to the serverless application paradigm. You can use the AWS Fault Injection Service (FIS) to automate and manage chaos experiments across different Lambda functions to provide a reusable testing method.

Amazon ECS and AWS Fargate

Amazon Elastic Container Service (Amazon ECS) now provides managed instance draining as a built-in feature of Amazon ECS capacity providers. This allows Amazon ECS to safely and automatically drain tasks from Amazon Elastic Compute Cloud (Amazon EC2) instances that are part of an Amazon EC2 Auto Scaling Group associated with an Amazon ECS capacity provider. This simplification allows you to remove custom lifecycle hooks previously used to drain Amazon EC2 instances. You can now perform infrastructure updates such as rolling out a new version of the ECS agent by seamlessly using Auto Scaling Group instance refresh, with Amazon ECS ensuring workloads are not interrupted.

Credentials Fetcher makes it easier to run containers that depend on Windows authentication when using Amazon EC2. Credentials Fetcher now integrates with Amazon ECS, using either the Amazon EC2 launch type, or AWS Fargate serverless compute launch type.

Amazon ECS Service Connect is a networking capability to simplify service discovery, connectivity, and traffic observability for Amazon ECS. You can now more easily integrate certificate management to encrypt service-to-service communication using Transport Layer Security (TLS). You do not need to modify your application code, add additional network infrastructure, or operate service mesh solutions.

Amazon ECS Service Connect

Amazon ECS Service Connect

Running distributed machine learning (ML) workloads on Amazon ECS allows ML teams to focus on creating, training and deploying models, rather than spending time managing the container orchestration engine. Amazon ECS provides a great environment to run ML projects as it supports workloads that use NVIDIA GPUs and provides optimized images with pre-installed NVIDIA Kernel drivers and Docker runtime.

See how to build preview environments for Amazon ECS applications with AWS Copilot. AWS Copilot is an open source command line interface that makes it easier to build, release, and operate production ready containerized applications.

Learn techniques for automatic scaling of your Amazon Elastic Container Service  (Amazon ECS) container workloads to enhance the end user experience. This post explains how to use AWS Application Auto Scaling which helps you configure automatic scaling of your Amazon ECS service. You can also use Amazon ECS Service Connect and AWS Distro for OpenTelemetry (ADOT) in Application Auto Scaling.

AWS Step Functions

AWS workloads sometimes require access to data stored in on-premises databases and storage locations. Traditional solutions to establish connectivity to the on-premises resources require inbound rules to firewalls, a VPN tunnel, or public endpoints. Discover how to use the MQTT protocol (AWS IoT Core) with AWS Step Functions to dispatch jobs to on-premises workers to access or retrieve data stored on-premises.

You can use Step Functions to orchestrate many business processes. Many industries are required to provide audit trails for decision and transactional systems. Learn how to build a serverless pipeline to create a reliable, performant, traceable, and durable pipeline for audit processing.

Amazon EventBridge

Amazon EventBridge now supports publishing events to AWS AppSync GraphQL APIs as native targets. The new integration allows you to publish events easily to a wider variety of consumers and simplifies updating clients with near real-time data.

Amazon EventBridge publishing events to AWS AppSync

Amazon EventBridge publishing events to AWS AppSync

Discover how to send and receive CloudEvents with EventBridge. CloudEvents is an open-source specification for describing event data in a common way. You can publish CloudEvents directly to EventBridge, filter and route them, and use input transformers and API Destinations to send CloudEvents to downstream AWS services and third-party APIs.

AWS Application Composer

AWS Application Composer lets you create infrastructure as code templates by dragging and dropping cards on a virtual canvas. These represent CloudFormation resources, which you can wire together to create permissions and references. Application Composer has now expanded to the VS Code IDE as part of the AWS Toolkit. This now includes a generative AI partner that helps you write infrastructure as code (IaC) for all 1100+ AWS CloudFormation resources that Application Composer now supports.

AWS AppComposer generate suggestions

AWS AppComposer generate suggestions

Amazon API Gateway

Learn how to consume private Amazon API Gateway APIs using mutual TLS (mTLS). mTLS helps prevent man-in-the-middle attacks and protects against threats such as impersonation attempts, data interception, and tampering.

Serverless at AWS re:Invent

Serverless at AWS reInvent

Serverless at AWS reInvent

Visit the Serverless Land YouTube channel to find a list of serverless and serverless container sessions from reinvent 2023. Hear from experts like Chris Munns and Julian Wood in their popular session, Best practices for serverless developers, or Nathan Peck and Jessica Deen in Deploying multi-tenant SaaS applications on Amazon ECS and AWS Fargate.

Serverless blog posts

January

February

March

Serverless container blog posts

January

February

December

Serverless Office Hours

Serverless Office Hours

Serverless Office Hours

January

February

March

Containers from the Couch

Containers from the Couch

Containers from the Couch

January

February

March

FooBar Serverless

FooBar Serverless

FooBar Serverless

January

February

March

Still looking for more?

The Serverless landing page has more information. The Lambda resources page contains case studies, webinars, whitepapers, customer stories, reference architectures, and even more Getting Started tutorials.

You can also follow the Serverless Developer Advocacy team on Twitter to see the latest news, follow conversations, and interact with the team.

And finally, visit the Serverless Land and Containers on AWS websites for all your serverless and serverless container needs.

AWS Weekly Roundup — Claude 3 Sonnet support in Bedrock, new instances, and more — March 11, 2024

Post Syndicated from Marcia Villalba original https://aws.amazon.com/blogs/aws/aws-weekly-roundup-claude-3-sonnet-support-in-bedrock-new-instances-and-more-march-11-2024/

Last Friday was International Women’s Day (IWD), and I want to take a moment to appreciate the amazing ladies in the cloud computing space that are breaking the glass ceiling by reaching technical leadership positions and inspiring others to go and build, as our CTO Werner Vogels says.Now go build

Last week’s launches
Here are some launches that got my attention during the previous week.

Amazon Bedrock – Now supports Anthropic’s Claude 3 Sonnet foundational model. Claude 3 Sonnet is two times faster and has the same level of intelligence as Anthropic’s highest-performing models, Claude 2 and Claude 2.1. My favorite characteristic is that Sonnet is better at producing JSON outputs, making it simpler for developers to build applications. It also offers vision capabilities. You can learn more about this foundation model (FM) in the post that Channy wrote early last week.

AWS re:Post – Launched last week! AWS re:Post Live is a weekly Twitch livestream show that provides a way for the community to reach out to experts, ask questions, and improve their skills. The show livestreams every Monday at 11 AM PT.

Amazon CloudWatchNow streams daily metrics on CloudWatch metric streams. You can use metric streams to send a stream of near real-time metrics to a destination of your choice.

Amazon Elastic Compute Cloud (Amazon EC2)Announced the general availability of new metal instances, C7gd, M7gd, and R7gd. These instances have up to 3.8 TB of local NVMe-based SSD block-level storage and are built on top of the AWS Nitro System.

AWS WAFNow supports configurable evaluation time windows for request aggregation with rate-based rules. Previously, AWS WAF was fixed to a 5-minute window when aggregating and evaluating the rules. Now you can select windows of 1, 2, 5 or 10 minutes, depending on your application use case.

AWS Partners – Last week, we announced the AWS Generative AI Competency Partners. This new specialization features AWS Partners that have shown technical proficiency and a track record of successful projects with generative artificial intelligence (AI) powered by AWS.

For a full list of AWS announcements, be sure to keep an eye on the What’s New at AWS page.

Other AWS news
Some other updates and news that you may have missed:

One of the articles that caught my attention recently compares different design approaches for building serverless microservices. This article, written by Luca Mezzalira and Matt Diamond, compares the three most common designs for serverless workloads and explains the benefits and challenges of using one over the other.

And if you are interested in the serverless space, you shouldn’t miss the Serverless Office Hours, which airs live every Tuesday at 10 AM PT. Join the AWS Serverless Developer Advocates for a weekly chat on the latest from the serverless space.

Serverless office hours

The Official AWS Podcast – Listen each week for updates on the latest AWS news and deep dives into exciting use cases. There are also official AWS podcasts in several languages. Check out the ones in FrenchGermanItalian, and Spanish.

AWS Open Source News and Updates – This is a newsletter curated by my colleague Ricardo to bring you the latest open source projects, posts, events, and more.

Upcoming AWS events
Check your calendars and sign up for these AWS events:

AWS Summit season is about to start. The first ones are Paris (April 3), Amsterdam (April 9), and London (April 24). AWS Summits are free events that you can attend in person and learn about the latest in AWS technology.

GOTO x AWS EDA Day London 2024 – On May 14, AWS partners with GOTO bring to you the event-driven architecture (EDA) day conference. At this conference, you will get to meet experts in the EDA space and listen to very interesting talks from customers, experts, and AWS.

GOTO EDA Day 2022

You can browse all upcoming in-person and virtual events here.

That’s all for this week. Check back next Monday for another Week in Review!

— Marcia

This post is part of our Weekly Roundup series. Check back each week for a quick roundup of interesting news and announcements from AWS!

Top Architecture Blog Posts of 2023

Post Syndicated from Andrea Courtright original https://aws.amazon.com/blogs/architecture/top-architecture-blog-posts-of-2023/

2023 was a rollercoaster year in tech, and we at the AWS Architecture Blog feel so fortunate to have shared in the excitement. As we move into 2024 and all of the new technologies we could see, we want to take a moment to highlight the brightest stars from 2023.

As always, thanks to our readers and to the many talented and hardworking Solutions Architects and other contributors to our blog.

I give you our 2023 cream of the crop!

#10: Build a serverless retail solution for endless aisle on AWS

In this post, Sandeep and Shashank help retailers and their customers alike in this guided approach to finding inventory that doesn’t live on shelves.

Building endless aisle architecture for order processing

Figure 1. Building endless aisle architecture for order processing

Check it out!

#9: Optimizing data with automated intelligent document processing solutions

Who else dreads wading through large amounts of data in multiple formats? Just me? I didn’t think so. Using Amazon AI/ML and content-reading services, Deependra, Anirudha, Bhajandeep, and Senaka have created a solution that is scalable and cost-effective to help you extract the data you need and store it in a format that works for you.

AI-based intelligent document processing engine

Figure 2: AI-based intelligent document processing engine

Check it out!

#8: Disaster Recovery Solutions with AWS managed services, Part 3: Multi-Site Active/Passive

Disaster recovery posts are always popular, and this post by Brent and Dhruv is no exception. Their creative approach in part 3 of this series is most helpful for customers who have business-critical workloads with higher availability requirements.

Warm standby with managed services

Figure 3. Warm standby with managed services

Check it out!

#7: Simulating Kubernetes-workload AZ failures with AWS Fault Injection Simulator

Continuing with the theme of “when bad things happen,” we have Siva, Elamaran, and Re’s post about preparing for workload failures. If resiliency is a concern (and it really should be), the secret is test, test, TEST.

Architecture flow for Microservices to simulate a realistic failure scenario

Figure 4. Architecture flow for Microservices to simulate a realistic failure scenario

Check it out!

#6: Let’s Architect! Designing event-driven architectures

Luca, Laura, Vittorio, and Zamira weren’t content with their four top-10 spots last year – they’re back with some things you definitely need to know about event-driven architectures.

Let's Architect

Figure 5. Let’s Architect artwork

Check it out!

#5: Use a reusable ETL framework in your AWS lake house architecture

As your lake house increases in size and complexity, you could find yourself facing maintenance challenges, and Ashutosh and Prantik have a solution: frameworks! The reusable ETL template with AWS Glue templates might just save you a headache or three.

Reusable ETL framework architecture

Figure 6. Reusable ETL framework architecture

Check it out!

#4: Invoking asynchronous external APIs with AWS Step Functions

It’s possible that AWS’ menagerie of services doesn’t have everything you need to run your organization. (Possible, but not likely; we have a lot of amazing services.) If you are using third-party APIs, then Jorge, Hossam, and Shirisha’s architecture can help you maintain a secure, reliable, and cost-effective relationship among all involved.

Invoking Asynchronous External APIs architecture

Figure 7. Invoking Asynchronous External APIs architecture

Check it out!

#3: Announcing updates to the AWS Well-Architected Framework

The Well-Architected Framework continues to help AWS customers evaluate their architectures against its six pillars. They are constantly striving for improvement, and Haleh’s diligence in keeping us up to date has not gone unnoticed. Thank you, Haleh!

Well-Architected logo

Figure 8. Well-Architected logo

Check it out!

#2: Let’s Architect! Designing architectures for multi-tenancy

The practically award-winning Let’s Architect! series strikes again! This time, Luca, Laura, Vittorio, and Zamira were joined by Federica to discuss multi-tenancy and why that concept is so crucial for SaaS providers.

Let's Architect

Figure 9. Let’s Architect

Check it out!

And finally…

#1: Understand resiliency patterns and trade-offs to architect efficiently in the cloud

Haresh, Lewis, and Bonnie revamped this 2022 post into a masterpiece that completely stole our readers’ hearts and is among the top posts we’ve ever made!

Resilience patterns and trade-offs

Figure 10. Resilience patterns and trade-offs

Check it out!

Bonus! Three older special mentions

These three posts were published before 2023, but we think they deserve another round of applause because you, our readers, keep coming back to them.

Thanks again to everyone for their contributions during a wild year. We hope you’re looking forward to the rest of 2024 as much as we are!

Disaster recovery strategies for Amazon MWAA – Part 1

Post Syndicated from Parnab Basak original https://aws.amazon.com/blogs/big-data/disaster-recovery-strategies-for-amazon-mwaa-part-1/

In the dynamic world of cloud computing, ensuring the resilience and availability of critical applications is paramount. Disaster recovery (DR) is the process by which an organization anticipates and addresses technology-related disasters. For organizations implementing critical workload orchestration using Amazon Managed Workflows for Apache Airflow (Amazon MWAA), it is crucial to have a DR plan in place to ensure business continuity.

In this series, we explore the need for Amazon MWAA disaster recovery and prescribe solutions that will sustain Amazon MWAA environments against unintended disruptions. This lets you to define, avoid, and handle disruption risks as part of your business continuity plan. This post focuses on designing the overall DR architecture. A future post in this series will focus on implementing the individual components using AWS services.

The need for Amazon MWAA disaster recovery

Amazon MWAA, a fully managed service for Apache Airflow, brings immense value to organizations by automating workflow orchestration for extract, transform, and load (ETL), DevOps, and machine learning (ML) workloads. Amazon MWAA has a distributed architecture with multiple components such as scheduler, worker, web server, queue, and database. This makes it difficult to implement a comprehensive DR strategy.

An active Amazon MWAA environment continuously parses Airflow Directed Acyclic Graphs (DAGs), reading them from a configured Amazon Simple Storage Service (Amazon S3) bucket. DAG source unavailability due to network unreachability, unintended corruption, or deletes leads to extended downtime and service disruption.

Within Airflow, the metadata database is a core component storing configuration variables, roles, permissions, and DAG run histories. A healthy metadata database is therefore critical for your Airflow environment. As with any core Airflow component, having a backup and disaster recovery plan in place for the metadata database is essential.

Amazon MWAA deploys Airflow components to multiple Availability Zones within your VPC in your preferred AWS Region. This provides fault tolerance and automatic recovery against a single Availability Zone failure. For mission-critical workloads, being resilient to the impairments of a unitary Region through multi-Region deployments is additionally important to ensure high availability and business continuity.

Balancing between costs to maintain redundant infrastructures, complexity, and recovery time is essential for Amazon MWAA environments. Organizations aim for cost-effective solutions that minimize their Recovery Time Objective (RTO) and Recovery Point Objective (RPO) to meet their service level agreements, be economically viable, and meet their customers’ demands.

Detect disasters in the primary environment: Proactive monitoring through metrics and alarms

Prompt detection of disasters in the primary environment is crucial for timely disaster recovery. Monitoring the Amazon CloudWatch SchedulerHeartbeat metric provides insights into Airflow health of an active Amazon MWAA environment. You can add other health check metrics to the evaluation criteria, such as checking the availability of upstream or downstream systems and network reachability. Combined with CloudWatch alarms, you can send notifications when these thresholds over a number of time periods are not met. You can add alarms to dashboards to monitor and receive alerts about your AWS resources and applications across multiple Regions.

AWS publishes our most up-to-the-minute information on service availability on the Service Health Dashboard. You can check at any time to get current status information, or subscribe to an RSS feed to be notified of interruptions to each individual service in your operating Region. The AWS Health Dashboard provides information about AWS Health events that can affect your account.

By combining metric monitoring, available dashboards, and automatic alarming, you can promptly detect unavailability of your primary environment, enabling proactive measures to transition to your DR plan. It is critical to factor in incident detection, notification, escalation, discovery, and declaration into your DR planning and implementation to provide realistic and achievable objectives that provide business value.

In the following sections, we discuss two Amazon MWAA DR strategy solutions and their architecture.

DR strategy solution 1: Backup and restore

The backup and restore strategy involves generating Airflow component backups in the same or different Region as your primary Amazon MWAA environment. To ensure continuity, you can asynchronously replicate these to your DR Region, with minimal performance impact on your primary Amazon MWAA environment. In the event of a rare primary Regional impairment or service disruption, this strategy will create a new Amazon MWAA environment and recover historical data to it from existing backups. However, it’s important to note that during the recovery process, there will be a period where no Airflow environments are operational to process workflows until the new environment is fully provisioned and marked as available.

This strategy provides a low-cost and low-complexity solution that is also suitable for mitigating against data loss or corruption within your primary Region. The amount of data being backed up and the time to create a new Amazon MWAA environment (typically 20–30 minutes) affects how quickly restoration can happen. To enable infrastructure to be redeployed quickly without errors, deploy using infrastructure as code (IaC). Without IaC, it may be complex to restore an analogous DR environment, which will lead to increased recovery times and possibly exceed your RTO.

Let’s explore the setup required when your primary Amazon MWAA environment is actively running, as shown in the following figure.

Backup and Restore - Pre

The solution comprises three key components. The first component is the primary environment, where the Airflow workflows are initially deployed and actively running. The second component is the disaster monitoring component, comprised of CloudWatch and a combination of an AWS Step Functions state machine and a AWS Lambda function. The third component is for creating and storing backups of all configurations and metadata that is required to restore. This can be in the same Region as your primary or replicated to your DR Region using S3 Cross-Region Replication (CRR). For CRR, you also pay for inter-Region data transfer out from Amazon S3 to each destination Region.

The first three steps in the workflow are as follows:

  1. As part of your backup creation process, Airflow metadata is replicated to an S3 bucket using an export DAG utility, run periodically based on your RPO interval.
  2. Your existing primary Amazon MWAA environment automatically emits the status of its scheduler’s health to the CloudWatch SchedulerHeartbeat metric.
  3. A multi-step Step Functions state machine is triggered from a periodic Amazon EventBridge schedule to monitor the scheduler’s health status. As the primary step of the state machine, a Lambda function evaluates the status of the SchedulerHeartbeat metric. If the metric is deemed healthy, no action is taken.

The following figure illustrates the additional steps in the solution workflow.

Backup and Restore post

  1. When the heartbeat count deviates from the normal count for a period of time, a series of actions are initiated to recover to a new Amazon MWAA environment in the DR Region. These actions include starting creation of a new Amazon MWAA environment, replicating the primary environment configurations, and then waiting for the new environment to become available.
  2. When the environment is available, an import DAG utility is run to restore the metadata contents from the backups. Any DAG runs that were interrupted during the impairment of the primary environment need to be manually rerun to maintain service level agreements. Future DAG runs are queued to run as per their next configured schedule.

DR strategy solution 2: Active-passive environments with periodic data synchronization

The active-passive environments with periodic data synchronization strategy focuses on maintaining recurrent data synchronization between an active primary and a passive Amazon MWAA DR environment. By periodically updating and synchronizing DAG stores and metadata databases, this strategy ensures that the DR environment remains current or nearly current with the primary. The DR Region can be the same or a different Region than your primary Amazon MWAA environment. In the event of a disaster, backups are available to revert to a previous known good state to minimize data loss or corruption.

This strategy provides low RTO and RPO with frequent synchronization, allowing quick recovery with minimal data loss. The infrastructure costs and code deployments are compounded to maintain both the primary and DR Amazon MWAA environments. Your DR environment is available immediately to run DAGs on.

The following figure illustrates the setup required when your primary Amazon MWAA environment is actively running.

Active Passive pre

The solution comprises four key components. Similar to the backup and restore solution, the first component is the primary environment, where the workflow is initially deployed and is actively running. The second component is the disaster monitoring component, consisting of CloudWatch and a combination of a Step Functions state machine and Lambda function. The third component creates and stores backups for all configurations and metadata required for the database synchronization. This can be in the same Region as your primary or replicated to your DR Region using Amazon S3 Cross-Region Replication. As mentioned earlier, for CRR, you also pay for inter-Region data transfer out from Amazon S3 to each destination Region. The last component is a passive Amazon MWAA environment that has the same Airflow code and environment configurations as the primary. The DAGs are deployed in the DR environment using the same continuous integration and continuous delivery (CI/CD) pipeline as the primary. Unlike the primary, DAGs are kept in a paused state to not cause duplicate runs.

The first steps of the workflow are similar to the backup and restore strategy:

  1. As part of your backup creation process, Airflow metadata is replicated to an S3 bucket using an export DAG utility, run periodically based on your RPO interval.
  2. Your existing primary Amazon MWAA environment automatically emits the status of its scheduler’s health to CloudWatch SchedulerHeartbeat metric.
  3. A multi-step Step Functions state machine is triggered from a periodic Amazon EventBridge schedule to monitor scheduler health status. As the primary step of the state machine, a Lambda function evaluates the status of the SchedulerHeartbeat metric. If the metric is deemed healthy, no action is taken.

The following figure illustrates the final steps of the workflow.

Active Passive post

  1. When the heartbeat count deviates from the normal count for a period of time, DR actions are initiated.
  2. As a first step, a Lambda function triggers an import DAG utility to restore the metadata contents from the backups to the passive Amazon MWAA DR environment. When the imports are complete, the same DAG can un-pause the other Airflow DAGs, making them active for future runs. Any DAG runs that were interrupted during the impairment of the primary environment need to be manually rerun to maintain service level agreements. Future DAG runs are queued to run as per their next configured schedule.

Best practices to improve resiliency of Amazon MWAA

To enhance the resiliency of your Amazon MWAA environment and ensure smooth disaster recovery, consider implementing the following best practices:

  • Robust backup and restore mechanisms – Implementing comprehensive backup and restore mechanisms for Amazon MWAA data is essential. Regularly deleting existing metadata based on your organization’s retention policies reduces backup times and makes your Amazon MWAA environment more performant.
  • Automation using IaC – Using automation and orchestration tools such as AWS CloudFormation, the AWS Cloud Development Kit (AWS CDK), or Terraform can streamline the deployment and configuration management of Amazon MWAA environments. This ensures consistency, reproducibility, and faster recovery during DR scenarios.
  • Idempotent DAGs and tasks – In Airflow, a DAG is considered idempotent if rerunning the same DAG with the same inputs multiple times has the same effect as running it only once. Designing idempotent DAGs and keeping tasks atomic decreases recovery time from failures when you have to manually rerun an interrupted DAG in your recovered environment.
  • Regular testing and validation – A robust Amazon MWAA DR strategy should include regular testing and validation exercises. By simulating disaster scenarios, you can identify any gaps in your DR plans, fine-tune processes, and ensure your Amazon MWAA environments are fully recoverable.

Conclusion

In this post, we explored the challenges for Amazon MWAA disaster recovery and discussed best practices to improve resiliency. We examined two DR strategy solutions: backup and restore and active-passive environments with periodic data synchronization. By implementing these solutions and following best practices, you can protect your Amazon MWAA environments, minimize downtime, and mitigate the impact of disasters. Regular testing, validation, and adaptation to evolving requirements are crucial for an effective Amazon MWAA DR strategy. By continuously evaluating and refining your disaster recovery plans, you can ensure the resilience and uninterrupted operation of your Amazon MWAA environments, even in the face of unforeseen events.

For additional details and code examples on Amazon MWAA, refer to the Amazon MWAA User Guide and the Amazon MWAA examples GitHub repo.


About the Authors

Parnab Basak is a Senior Solutions Architect and a Serverless Specialist at AWS. He specializes in creating new solutions that are cloud native using modern software development practices like serverless, DevOps, and analytics. Parnab works closely in the analytics and integration services space helping customers adopt AWS services for their workflow orchestration needs.

Chandan Rupakheti is a Solutions Architect and a Serverless Specialist at AWS. He is a passionate technical leader, researcher, and mentor with a knack for building innovative solutions in the cloud and bringing stakeholders together in their cloud journey. Outside his professional life, he loves spending time with his family and friends besides listening and playing music.

Vinod Jayendra is a Enterprise Support Lead in ISV accounts at Amazon Web Services, where he helps customers in solving their architectural, operational, and cost optimization challenges. With a particular focus on Serverless technologies, he draws from his extensive background in application development to deliver top-tier solutions. Beyond work, he finds joy in quality family time, embarking on biking adventures, and coaching youth sports team.

Rupesh Tiwari is a Senior Solutions Architect at AWS in New York City, with a focus on Financial Services. He has over 18 years of IT experience in the finance, insurance, and education domains, and specializes in architecting large-scale applications and cloud-native big data workloads. In his spare time, Rupesh enjoys singing karaoke, watching comedy TV series, and creating joyful moments with his family.

Enable metric-based and scheduled scaling for Amazon Managed Service for Apache Flink

Post Syndicated from Francisco Morillo original https://aws.amazon.com/blogs/big-data/enable-metric-based-and-scheduled-scaling-for-amazon-managed-service-for-apache-flink/

Thousands of developers use Apache Flink to build streaming applications to transform and analyze data in real time. Apache Flink is an open source framework and engine for processing data streams. It’s highly available and scalable, delivering high throughput and low latency for the most demanding stream-processing applications. Monitoring and scaling your applications is critical to keep your applications running successfully in a production environment.

Amazon Managed Service for Apache Flink is a fully managed service that reduces the complexity of building and managing Apache Flink applications. Amazon Managed Service for Apache Flink manages the underlying Apache Flink components that provide durable application state, metrics, logs, and more.

In this post, we show a simplified way to automatically scale up and down the number of KPUs (Kinesis Processing Units; 1 KPU is 1 vCPU and 4 GB of memory) of your Apache Flink applications with Amazon Managed Service for Apache Flink. We show you how to scale by using metrics such as CPU, memory, backpressure, or any custom metric of your choice. Additionally, we show how to perform scheduled scaling, allowing you to adjust your application’s capacity at specific times, particularly when dealing with predictable workloads. We also share an AWS CloudFormation utility to help you implement auto scaling quickly with your Amazon Managed Service for Apache Flink applications.

Metric-based scaling

This section describes how to implement a scaling solution for Amazon Managed Service for Apache Flink based on Amazon CloudWatch metrics. Amazon Managed Service for Apache Flink comes with an auto scaling option out of the box that scales out when container CPU utilization is above 75% for 15 minutes. This works well for many use cases; however, for some applications, you may need to scale based on a different metric, or trigger the scaling action at a certain point in time or by a different factor. You can customize your scaling policies and save costs by right-sizing your Amazon Managed Apache Flink applications the deploying this solution.

To perform metric-based scaling, we use CloudWatch alarms, Amazon EventBridge, AWS Step Functions, and AWS Lambda. You can choose from metrics coming from the source such as Amazon Kinesis Data Streams or Amazon Managed Streaming for Apache Kafka (Amazon MSK), or metrics from the Amazon Managed Service for Apache Flink application. You can find these components in the CloudFormation template in the GitHub repo.

The following diagram shows how to scale an Amazon Managed Service for Apache Flink application in response to a CloudWatch alarm.

This solution uses the metric selected and creates two CloudWatch alarms that, depending on the threshold you use, trigger a rule in EventBridge to start running a Step Functions state machine. The following diagram illustrates the state machine workflow.

Note: Amazon Kinesis Data Analytics was renamed to Amazon Managed Service for Apache Flink August 2023

The Step Functions workflow consists of the following steps:

  1. The state machine describes the Amazon Managed Service for Apache Flink application, which will provide information related to the current number of KPUs in the application, as well if the application is being updated or is it running.
  2. The state machine invokes a Lambda function that, depending on which alarm was triggered, will scale the application up or down, following the parameters set in the CloudFormation template. When scaling the application, it will use the increase factor (either add/subtract or multiple/divide based on that factor) defined in the CloudFormation template. You can have different factors for scaling in or out. If you want to take a more cautious approach to scaling, you can use add/subtract and use an increase factor for scaling in/out of 1.
  3. If the application has reached the maximum or minimum number of KPUs set in the parameters of the CloudFormation template, the workflow stops. Keep in mind that Amazon Managed Service for Apache Flink applications have a default maximum of 64 KPUs (you can request to increase this limit). Do not specify a maximum value above 64 KPUs if you have not requested to increase the quota, because the scaling solution will get stuck by failing to update.
  4. If the workflow continues, because the allocated KPUs haven’t reached the maximum or minimum values, the workflow will wait for a period of time you specify, and then describe the application and see if it has finished updating.
  5. The workflow will continue to wait until the application has finished updating. When the application is updated, the workflow will wait for a period of time you specify in the CloudFormation template, to allow the metric to fall within the threshold and have the CloudWatch rule change from ALARM state to OK.
  6. If the metric is still in ALARM state, the workflow will start again and continue to scale the application either up or down. If the metric is in OK state, the workflow will stop.

For applications that read from a Kinesis Data Streams source, you can use the metric millisBehindLatest. If using a Kafka source, you can use records lag max for scaling events. These metrics capture how far behind your application is from the head of the stream. You can also use a custom metric that you have registered in your Apache Flink applications.

The sample CloudFormation template allows you to select one of the following metrics:

  • Amazon Managed Service for Apache Flink application metrics – Requires an application name:
    • ContainerCPUUtilization – Overall percentage of CPU utilization across task manager containers in the Flink application cluster.
    • ContainerMemoryUtilization – Overall percentage of memory utilization across task manager containers in the Flink application cluster.
    • BusyTimeMsPerSecond – Time in milliseconds the application is busy (neither idle nor back pressured) per second.
    • BackPressuredTimeMsPerSecond – Time in milliseconds the application is back pressured per second.
    • LastCheckpointDuration – Time in milliseconds it took to complete the last checkpoint.
  • Kinesis Data Streams metrics – Requires the data stream name:
    • MillisBehindLatest – The number of milliseconds the consumer is behind the head of the stream, indicating how far behind the current time the consumer is.
    • IncomingRecords – The number of records successfully put to the Kinesis data stream over the specified time period. If no records are coming, this metric will be null and you won’t be able to scale down.
  • Amazon MSK metrics – Requires the cluster name, topic name, and consumer group name):
    • MaxOffsetLag – The maximum offset lag across all partitions in a topic.
    • SumOffsetLag – The aggregated offset lag for all the partitions in a topic.
    • EstimatedMaxTimeLag – The time estimate (in seconds) to drain MaxOffsetLag.
  • Custom metrics – Metrics you can define as part of your Apache Flink applications. Most common metrics are counters (continuously increase) or gauges (can be updated with last value). For this solution, you need to add the kinesisAnalytics dimension to the metric group. You also need to provide the custom metric name as a parameter in the CloudFormation template. If you need to use more dimensions in your custom metric, you need to modify the CloudWatch alarm so it’s able to use your specific metric. For more information on custom metrics, see Using Custom Metrics with Amazon Managed Service for Apache Flink.

The CloudFormation template deploys the resources as well as the auto scaling code. You only need to specify the name of the Amazon Managed Service for Apache Flink application, the metric to which you want to scale your application in or out, and the thresholds for triggering an alarm. The solution by default will use the average aggregation for metrics and a period duration of 60 seconds for each data point. You can configure the evaluation periods and data points to alarm when defining the CloudFormation template.

Scheduled scaling

This section describes how to implement a scaling solution for Amazon Managed Service for Apache Flink based on a schedule. To perform scheduled scaling, we use EventBridge and Lambda, as illustrated in the following figure.

These components are available in the CloudFormation template in the GitHub repo.

The EventBridge scheduler is triggered based on the parameters set when deploying the CloudFormation template. You define the KPU of the applications when running at peak times, as well as the KPU for non-peak times. The application runs with those KPU parameters depending on the time of day.

As with the previous example for metric-based scaling, the CloudFormation template deploys the resources and scaling code required. You only need to specify the name of the Amazon Managed Service for Apache Flink application and the schedule for the scaler to modify the application to the set number of KPUs.

Considerations for scaling Flink applications using metric-based or scheduled scaling

Be aware of the following when considering these solutions:

  • When scaling Amazon Managed Service for Apache Flink applications in or out, you can choose to either increase the overall application parallelism or modify the parallelism per KPU. The latter allows you to set the number of parallel tasks that can be scheduled per KPU. This sample only updates the overall parallelism, not the parallelism per KPU.
  • If SnapshotsEnabled is set to true in ApplicationSnapshotConfiguration, Amazon Managed Service for Apache Flink will automatically pause the application, take a snapshot, and then restore the application with the updated configuration whenever it is updated or scaled. This process may result in downtime for the application, depending on the state size, but there will be no data loss. When using metric-based scaling, you have to choose a minimum and a maximum threshold of KPU the application can have. Depending on by how much you perform the scaling, if the new desired KPU is bigger or lower than your thresholds, the solution will update the KPUs to be equal to your thresholds.
  • When using metric-based scaling, you also have to choose a cooling down period. This is the amount of time you want your application to wait after being updated, to see if the metric has gone from ALARM status to OK status. This value depends on how long are you willing to wait before another scaling event to occur.
  • With the metric-based scaling solution, you are limited to choosing the metrics that are listed in the CloudFormation template. However, you can modify the alarms to use any available metric in CloudWatch.
  • If your application is required to run without interruptions for periods of time, we recommend using scheduled scaling, to limit scaling to non-critical times.

Summary

In this post, we covered how you can enable custom scaling for Amazon Managed Service for Apache Flink applications using enhanced monitoring features from CloudWatch integrated with Step Functions and Lambda. We also showed how you can configure a schedule to scale an application using EventBridge. Both of these samples and many more can be found in the GitHub repo.


About the Authors

Deepthi Mohan is a Principal PMT on the Amazon Managed Service for Apache Flink team.

Francisco Morillo is a Streaming Solutions Architect at AWS. Francisco works with AWS customers, helping them design real-time analytics architectures using AWS services, supporting Amazon Managed Streaming for Apache Kafka (Amazon MSK) and Amazon Managed Service for Apache Flink.

Happy New Year! AWS Weekly Roundup – January 8, 2024

Post Syndicated from Channy Yun original https://aws.amazon.com/blogs/aws/happy-new-year-aws-weekly-roundup-january-8-2024/

Happy New Year! Cloud technologies, machine learning, and generative AI have become more accessible, impacting nearly every aspect of our lives. Amazon CTO Dr. Werner Vogels offers four tech predictions for 2024 and beyond:

  • Generative AI becomes culturally aware
  • FemTech finally takes off
  • AI assistants redefine developer productivity
  • Education evolves to match the speed of technology

Read how these technology trends will converge to help solve some of society’s most difficult problems. Download the Werner Vogels’ Tech Predictions for 2024 and Beyond ebook or read Werner’s All Things Distributed blog.

AWS re:Invent 2023To hear insights from AWS and industry thought leaders, grow your skills, and get inspired, watch AWS re:Invent 2023 videos on demand for keynotes, innovation talks, breakout sessions, and AWS Hero guide playlists.

Launches from the last few weeks
Since our last week in review on December 18, 2023, I’d like to highlight some launches from year end, as well as last week:

New AWS Canada West (Calgary) Region – We are opening a new and second Region and in Canada, AWS Canada West (Calgary). At the end of 2023, AWS had 33 AWS Regions and 105 Availability Zones (AZs) globally. We preannounced 12 additional AZs in four future Regions in Malaysia, New Zealand, Thailand, and the AWS European Sovereign Cloud. We will share more information on these Regions in 2024. Please stay tuned.

DNS over HTTPS in Amazon Route 53 Resolver – You can use the DNS over HTTPS (DoH) protocol for both inbound and outbound Route 53 Resolver endpoints. As the name suggests, DoH supports HTTP or HTTP/2 over TLS to encrypt the data exchanged for Domain Name System (DNS) resolutions.

Automatic enrollment to Amazon RDS Extended Support – Your MySQL 5.7 and PostgreSQL 11 database instances running on Amazon Aurora and Amazon RDS will be automatically enrolled into Amazon RDS Extended Support starting on February 29, 2024. You can have more control over when you want to upgrade the major version of your database after the community end of life (EoL).

New Amazon CloudWatch Network Monitor – This is a new feature of Amazon CloudWatch that helps monitor network availability and performance between AWS and your on-premises environments. Network Monitor needs zero manual instrumentation and gives you access to real-time network visibility to proactively and quickly identify issues within the AWS network and your own hybrid environment. For more information, read Monitor hybrid connectivity with Amazon CloudWatch Network Monitor.

Amazon Aurora PostgreSQL integrations with Amazon Bedrock – You can use two methods to integrate Aurora PostgreSQL databases with Amazon Bedrock to power generative AI applications. You can use the SQL query with Aurora ML integration with Amazon Bedrock and Aurora vector store with Knowledge Bases for Amazon Bedrock for Retrieval Augmented Generation (RAG).

New WordPress setup on Amazon Lightsail – Set up your WordPress website on Amazon Lightsail with the new workflow to eliminate complexity and time spent configuring your website. The workflow allows you to complete all the necessary steps, including setting up a Secure Sockets Layer (SSL) certificate to secure your website with HTTPS.

For a full list of AWS announcements, be sure to keep an eye on the What’s New at AWS page.

Other AWS News
Here are some other news items that you may find interesting in the new year:

Book recommendations for AWS customer executives – Plan for the new year and catch up on what others are doing and thinking. AWS Enterprise Strategy team recommends what books are most important for our AWS customer executives to read.

Best practices for scaling AWS CDK adoption with Platform Engineering – A recent evolution in DevOps is the introduction of platform engineering teams to build services, toolchains, and documentation to support workload teams. This blog post introduces strategies and best practices for accelerating CDK adoption within your organization. You can learn how to scale the lessons learned from the pilot project across your organization through platform engineering.

High performance running HPC applications on AWS Graviton instances – When running the Parallel Lattice Boltzmann Solver (Palabos) on Amazon EC2 Hpc7g instances to solve computational fluid dynamics (CFD) problems, performance increased by up to 70% and price performance was up to 3x better than on the previous generation of Graviton instances.

The new AWS open source newsletter, #181 – Check up on all the latest open source content, which this week includes AWS Amplify, Amazon Corretto, dbt, Apache Flink, Karpenter, LangChain, Pinecone, and more.

Upcoming AWS Events
Check your calendars and sign up for these AWS events in the new year:

AWS at CES 2024 (January 9-12) – AWS will be representing some of the latest cloud services and solutions that are purpose built for the automotive, mobility, transportation, and manufacturing industries. Join us to learn about the latest cloud capabilities across generative AI, software define vehicles, product engineering, sustainability, new digital customer experiences, connected mobility, autonomous driving, and so much more in Amazon Experience Area.

APJ Builders Online Series (January 18) – This online conference is designed for you to learn core AWS concepts, and step-by-step architectural best practices, including demonstrations to help you get started and accelerate your success on AWS.

You can browse all upcoming AWS-led in-person and virtual events, and developer-focused events such as AWS DevDay.

That’s all for this week. Check back next Monday for another Week in Review!

— Channy

This post is part of our Week in Review series. Check back each week for a quick roundup of interesting news and announcements from AWS!

AWS Weekly Roundup — AWS Lambda, AWS Amplify, Amazon OpenSearch Service, Amazon Rekognition, and more — December 18, 2023

Post Syndicated from Donnie Prakoso original https://aws.amazon.com/blogs/aws/aws-weekly-roundup-aws-lambda-aws-amplify-amazon-opensearch-service-amazon-rekognition-and-more-december-18-2023/

My memories of Amazon Web Services (AWS) re:Invent 2023 are still fresh even when I’m currently wrapping up my activities in Jakarta after participating in AWS Community Day Indonesia. It was a great experience, from delivering chalk talks and having thoughtful discussions with AWS service teams, to meeting with AWS Heroes, AWS Community Builders, and AWS User Group leaders. AWS re:Invent brings the global AWS community together to learn, connect, and be inspired by innovation. For me, that spirit of connection is what makes AWS re:Invent always special.

Here’s a quick look of my highlights at AWS re:Invent and AWS Community Day Indonesia:

If you missed AWS re:Invent, you can watch the keynotes and sessions on demand. Also, check out the AWS News Editorial Team’s Top announcements of AWS re:Invent 2023 for all the major launches.

Recent AWS launches
Here are some of the launches that caught my attention in the past two weeks:

Query MySQL and PostgreSQL with AWS Amplify – In this post, Channy wrote how you can now connect your MySQL and PostgreSQL databases to AWS Amplify with just a few clicks. It generates a GraphQL API to query your database tables using AWS CDK.

Migration Assistant for Amazon OpenSearch Service – With this self-service solution, you can smoothly migrate from your self-managed clusters to Amazon OpenSearch Service managed clusters or serverless collections.

AWS Lambda simplifies connectivity to Amazon RDS and RDS Proxy – Now you can connect your AWS Lambda to Amazon RDS or RDS proxy using the AWS Lambda console. With a guided workflow, this improvement helps to minimize complexities and efforts to quickly launch a database instance and correctly connect a Lambda function.

New no-code dashboard application to visualize IoT data – With this announcement, you can now visualize and interact with operational data from AWS IoT SiteWise using a new open source Internet of Things (IoT) dashboard.

Amazon Rekognition improves Face Liveness accuracy and user experience – This launch provides higher accuracy in detecting spoofed faces for your face-based authentication applications.

AWS Lambda supports additional concurrency metrics for improved quota monitoring – Add CloudWatch metrics for your Lambda quotas, to improve visibility into concurrency limits.

AWS Malaysia now supports 3D-Secure authentication – This launch enables 3DS2 transaction authentication required by banks and payment networks, facilitating your secure online payments.

Announcing AWS CloudFormation template generation for Amazon EventBridge Pipes – With this announcement, you can now streamline the deployment of your EventBridge resources with CloudFormation templates, accelerating event-driven architecture (EDA) development.

Enhanced data protection for CloudWatch Logs – With the enhanced data protection, CloudWatch Logs helps identify and redact sensitive data in your logs, preventing accidental exposure of personal data.

Send SMS via Amazon SNS in Asia Pacific – With this announcement, now you can use SMS messaging across Asia Pacific from the Jakarta Region.

Lambda adds support for Python 3.12 – This launch brings the latest Python version to your Lambda functions.

CloudWatch Synthetics upgrades Node.js runtime – Now you can use Node.js 16.1 runtimes for your canary functions.

Manage EBS Volumes for your EC2 fleets – This launch simplifies attaching and managing EBS volumes across your EC2 fleets.

See you next year!
This is the last AWS Weekly Roundup for this year, and we’d like to thank you for being our wonderful readers. We’ll be back to share more launches for you on January 8, 2024.

Happy holidays!

Donnie

Amazon CloudWatch Application Signals for automatic instrumentation of your applications (preview)

Post Syndicated from Veliswa Boya original https://aws.amazon.com/blogs/aws/amazon-cloudwatch-application-signals-for-automatic-instrumentation-of-your-applications-preview/

One of the challenges with distributed systems is that they are made up of many interdependent services, which add a degree of complexity when you are trying to monitor their performance. Determining which services and APIs are experiencing high latencies or degraded availability requires manually putting together telemetry signals. This can result in time and effort establishing the root cause of any issues with the system due to the inconsistent experiences across metrics, traces, logs, real user monitoring, and synthetic monitoring.

You want to provide your customers with continuously available and high-performing applications. At the same time, the monitoring that assures this must be efficient, cost-effective, and without undifferentiated heavy lifting.

Amazon CloudWatch Application Signals helps you automatically instrument applications based on best practices for application performance. There is no manual effort, no custom code, and no custom dashboards. You get a pre-built, standardized dashboard showing the most important metrics, such as volume of requests, availability, latency, and more, for the performance of your applications. In addition, you can define Service Level Objectives (SLOs) on your applications to monitor specific operations that matter most to your business. An example of an SLO could be to set a goal that a webpage should render within 2000 ms 99.9 percent of the time in a rolling 28-day interval.

Application Signals automatically correlates telemetry across metrics, traces, logs, real user monitoring, and synthetic monitoring to speed up troubleshooting and reduce application disruption. By providing an integrated experience for analyzing performance in the context of your applications, Application Signals gives you improved productivity with a focus on the applications that support your most critical business functions.

My personal favorite is the collaboration between teams that’s made possible by Application Signals. I started this post by mentioning that distributed systems are made up of many interdependent services. On the Service Map, which we will look at later in this post, if you, as a service owner, identify an issue that’s caused by another service, you can send a link to the owner of the other service to efficiently collaborate on the triage tasks.

Getting started with Application Signals
You can easily collect application and container telemetry when creating new Amazon EKS clusters in the Amazon EKS console by enabling the new Amazon CloudWatch Observability EKS add-on. Another option is to enable for existing Amazon EKS Clusters or other compute types directly in the Amazon CloudWatch console.

Create service map

After enabling Application Signals via the Amazon EKS add-on or Custom option for other compute types, Application Signals automatically discovers services and generates a standard set of application metrics such as volume of requests and latency spikes or availability drops for APIs and dependencies, to name a few.

Specify platform

All of the services discovered and their golden metrics (volume of requests, latency, faults and errors) are then automatically displayed on the Services page and the Service Map. The Service Map gives you a visual deep dive to evaluate the health of a service, its operations, dependencies, and all the call paths between an operation and a dependency.

Auto-generated map

The list of services that are enabled in Application Signals will also show in the services dashboard, along with operational metrics across all of your services and dependencies to easily spot anomalies. The Application column is auto-populated if the EKS cluster belongs to an application that’s tagged in AppRegistry. The Hosted In column automatically detects which EKS pod, cluster, or namespace combination the service requests are running in, and you can select one to go directly to Container Insights for detailed container metrics such as CPU or memory utilization, to name a few.

Team collaboration with Application Signals
Now, to expand on the team collaboration that I mentioned at the beginning of this post. Let’s say you consult the services dashboard to do sanity checks and you notice two SLO issues for one of your services named pet-clinic-frontend. Your company maintains a set of SLOs, and this is the view that you use to understand how the applications are performing against the objectives. For the services that are tagged in AppRegistry all teams have a central view of the definition and ownership of the application. Further navigation to the service map gives you even more details on the health of this service.

At this point you make the decision to send the link to thepet-clinic-frontendservice to Sarah whose details you found in the AppRegistry. Sarah is the person on-call for this service. The link allows you to efficiently collaborate with Sarah because it’s been curated to land directly on the triage view that is contextualized based on your discovery of the issue. Sarah notices that the POST /api/customer/owners latency has increased to 2k ms for a number of requests and as the service owner, dives deep to arrive at the root cause.

Clicking into the latency graph returns a correlated list of traces that correspond directly to the operation, metric, and moment in time, which helps Sarah to find the exact traces that may have led to the increase in latency.

Sarah uses Amazon CloudWatch Synthetics and Amazon CloudWatch RUM and has enabled the X-Ray active tracing integration to automatically see the list of relevant canaries and pages correlated to the service. This integrated view now helps Sarah gain multiple perspectives in the performance of the application and quickly troubleshoot anomalies in a single view.

Available now
Amazon CloudWatch Application Signals is available in preview and you can start using it today in the following AWS Regions: US East (N. Virginia), US East (Ohio), US West (Oregon), Europe (Ireland), Asia Pacific (Sydney), and Asia Pacific (Tokyo).

To learn more, visit the Amazon CloudWatch user guide. You can submit your questions to AWS re:Post for Amazon CloudWatch, or through your usual AWS Support contacts.

Veliswa

Use natural language to query Amazon CloudWatch logs and metrics (preview)

Post Syndicated from Danilo Poccia original https://aws.amazon.com/blogs/aws/use-natural-language-to-query-amazon-cloudwatch-logs-and-metrics-preview/

To make it easy to interact with your operational data, Amazon CloudWatch is introducing today natural language query generation for Logs and Metrics Insights. With this capability, powered by generative artificial intelligence (AI), you can describe in English the insights you are looking for, and a Logs or Metrics Insights query will be automatically generated.

This feature provides three main capabilities for CloudWatch Logs and Metrics Insights:

  • Generate new queries from a description or a question to help you get started easily.
  • Query explanation to help you learn the language including more advanced features.
  • Refine existing queries using guided iterations.

Let’s see how these work in practice with a few examples. I’ll cover logs first and then metrics.

Generate CloudWatch Logs Insights queries with natural language
In the CloudWatch console, I select Log Insights in the Logs section. I then select the log group of an AWS Lambda function that I want to investigate.

I choose the Query generator button to open a new Prompt field where I enter what I need using natural language:

Tell me the duration of the 10 slowest invocations

Then, I choose Generate new query. The following Log Insights query is automatically generated:

fields @timestamp, @requestId, @message, @logStream, @duration 
| filter @type = "REPORT" and @duration > 1000
| sort @duration desc
| limit 10

Console screenshot.

I choose Run query to see the results.

Console screenshot.

I find that now there’s too much information in the output. I prefer to see only the data I need, so I enter the following sentence in the Prompt and choose Update query.

Show only timestamps and latency

The query is updated based on my input and only the timestamp and duration are returned:

fields @timestamp, @duration 
| filter @type = "REPORT" and @duration > 1000
| sort @duration desc
| limit 10

I run the updated query and get a result that is easier for me to read.

Console screenshot.

Now, I want to know if there are any errors in the log. I enter this sentence in the Prompt and generate a new query:

Count the number of ERROR messages

As requested, the generated query is counting the messages that contain the ERROR string:

fields @message
| filter @message like /ERROR/
| stats count()

I run the query and find out that there are more errors than I expected. I need more information.

Console screenshot.

I use this prompt to update the query and get a better distribution of the errors:

Show the errors per hour

The updated query uses the bin() function to group the result in one hour intervals.

fields @timestamp, @message
| filter @message like /ERROR/
| stats count(*) by bin(1h)

Let’s see a more advanced query about memory usage. I select the log groups of a few Lambda functions and type:

Show invocations with the most over-provisioned memory grouped by log stream

Before generating the query, I choose the gear icon to toggle the options to include my prompt and an explanation as comment. Here’s the result (I split the explanation over multiple lines for readability):

# Show invocations with the most over-provisioned memory grouped by log stream

fields @logStream, @memorySize/1000/1000 as memoryMB, @maxMemoryUsed/1000/1000 as maxMemoryUsedMB, (@memorySize/1000/1000 - @maxMemoryUsed/1000/1000) as overProvisionedMB 
| stats max(overProvisionedMB) as maxOverProvisionedMB by @logStream 
| sort maxOverProvisionedMB desc

# This query finds the amount of over-provisioned memory for each log stream by
# calculating the difference between the provisioned and maximum memory used.
# It then groups the results by log stream and calculates the maximum
# over-provisioned memory for each log stream. Finally, it sorts the results
# in descending order by the maximum over-provisioned memory to show
# the log streams with the most over-provisioned memory.

Now, I have the information I need to understand these errors. On the other side, I also have EC2 workloads. How are those instances running? Let’s look at some metrics.

Generate CloudWatch Metrics Insights queries with natural language
In the CloudWatch console, I select All metrics in the Metrics section. Then, in the Query tab, I use the Editor. If you prefer, the Query generator is available also in the Builder.

I choose Query generator like before. Then, I enter what I need using plain English:

Which 10 EC2 instances have the highest CPU utilization?

I choose Generate new query and get a result using the Metrics Insights syntax.

SELECT AVG("CPUUtilization")
FROM SCHEMA("AWS/EC2", InstanceId)
GROUP BY InstanceId
ORDER BY AVG() DESC
LIMIT 10

To see the graph, I choose Run.

Console screenshot.

Well, it looks like my EC2 instances are not doing much. This result shows how those instances are using the CPU, but what about storage? I enter this in the prompt and choose Update query:

How about the most EBS writes?

The updated query replaces the average CPU utilization with the sum of bytes written to all EBS volumes attached to the instance. It keeps the limit to only show the top 10 results.

SELECT SUM("EBSWriteBytes")
FROM SCHEMA("AWS/EC2", InstanceId)
GROUP BY InstanceId
ORDER BY SUM() DESC
LIMIT 10

I run the query and, by looking at the result, I have a better understanding of how storage is being used by my EC2 instances.

Try entering some requests and run the generated queries over your logs and metrics to see how this works with your data.

Things to know
Amazon CloudWatch natural language query generation for logs and metrics is available in preview in the US East (N. Virginia) and US West (Oregon) AWS Regions.

There is no additional cost for using natural language query generation during the preview. You only pay for the cost of running the queries according to CloudWatch pricing.

Generated queries are produced by generative AI and dependent on factors including the data selected and available in your account. For these reasons, your results may vary.

When generating a query, you can include your original request and an explanation of the query as comments. To do so, choose the gear icon in the bottom right corner of the query edit window and toggle those options.

This new capability can help you generate and update queries for logs and metrics, saving you time and effort. This approach allows engineering teams to scale their operations without worrying about specific data knowledge or query expertise.

Use natural language to analyze your logs and metrics with Amazon CloudWatch.

Danilo

Use Amazon CloudWatch to consolidate hybrid, multicloud, and on-premises metrics

Post Syndicated from Jeff Barr original https://aws.amazon.com/blogs/aws/new-use-amazon-cloudwatch-to-consolidate-hybrid-multi-cloud-and-on-premises-metrics/

You can now consolidate metrics from your hybrid, multicloud, and on-premises data sources using Amazon CloudWatch and process them in a consistent, unified fashion. You can query, visualize, and alarm on any and all of the metrics, regardless of their source. In addition to giving you a unified view, this new feature will help you to identify trends and issues that span multiple parts and aspects of your infrastructure.

When I first heard about this new feature, I thought, “Wait, I can do that with PutMetricData, what’s the big deal?” Quite a bit, as it turns out. PutMetricData stores the metrics in CloudWatch, but this cool new feature fetches them on demand, directly from the source.

Instead of storing data, you select and configure connectors that pull data from Amazon Managed Service for Prometheus, generic Prometheus, Amazon OpenSearch Service, Amazon RDS for MySQL, Amazon RDS for PostgreSQL, CSV files stored in Amazon Simple Storage Service (Amazon S3), and Microsoft Azure Monitor. Each connector is a AWS Lambda function that is deployed from a AWS CloudFormation template. CloudWatch invokes the appropriate Lambda functions as needed and makes use of the returned metrics immediately — they are not buffered or kept around.

Creating and using connectors
To get started I open the CloudWatch Console, click All metrics, and activate the Multi source query tab, then I click Create and manage data sources:

And then I do it again:

Then I choose a data source type:

CloudWatch will then prompt me for the details that it needs to create and set up the connector for my data source. For example, if I select Amazon RDS – MySQL, I give my data source a name, choose the RDS database instance, and specify the connection info:

When I click Create data source, a Lambda function, a Lambda Permission, an IAM role, a Secrets Manager Secret, a Log Group, and a AWS CloudFormation Stack will be created in my account:

Then, when I am ready to reference the data source and make use of the metrics that it provides, I enter a SQL query that returns timestamps and values for the metric:

Inside the Lambda function
The code for the Custom – getting started template is short, simple, and easy to understand. It implements handlers for two events:

DescribeGetMetricData – This handler returns a string that includes the name of the connector, default values for the arguments to the other handler, and a text description in Markdown format that is displayed in the custom data source query builder in the CloudWatch console.

GetMetricData – This handler returns a metric name, 1-dimensional array of timestamps and metric values, all for a time range that is provided as arguments to the handler.

If you spend a few minutes examining this code you should be able to see how to write functions to connect to your own data sources.

Things to know
Here are a couple of things to keep in mind about this powerful new feature:

Regions – You can create and use data connectors in all commercial AWS Regions; a connector that is running in one region can connect to and retrieve data from services and endpoints in other regions and other AWS accounts.

Pricing – There is no extra charge for the connectors. You pay for the invocations of the Lambda functions and for any other AWS infrastructure that you create.

Jeff;

Amazon CloudWatch Logs now offers automated pattern analytics and anomaly detection

Post Syndicated from Danilo Poccia original https://aws.amazon.com/blogs/aws/amazon-cloudwatch-logs-now-offers-automated-pattern-analytics-and-anomaly-detection/

Searching through log data to find operational or business insights often feels like looking for a needle in a haystack. It usually requires you to manually filter and review individual log records. To help you with that, Amazon CloudWatch has added new capabilities to automatically recognize and cluster patterns among log records, extract noteworthy content and trends, and notify you of anomalies using advanced machine learning (ML) algorithms trained using decades of Amazon and AWS operational data.

Specifically, CloudWatch now offers the following:

  • The Patterns tab on the Logs Insights page finds recurring patterns in your query results and lets you analyze them in detail. This makes it easier to find what you’re looking for and drill down into new or unexpected content in your logs.
  • The Compare button in the time interval selector on the Logs Insights page lets you quickly compare the query result for the selected time range to a previous period, such as the previous day, week, or month. In this way, it takes less time to see what has changed compared to a previous stable scenario.
  • The Log Anomalies page in the Logs section of the navigation pane automatically surfaces anomalies found in your logs while they are processed during ingestion.

Let’s see how these work in practice with a typical troubleshooting journey. I will look at some application logs to find key patterns, compare two time periods to understand what changed, and finally see how detecting anomalies can help discover issues.

Finding recurring patterns in the logs
In the CloudWatch console, I choose Logs Insights from the Logs section of the navigation pane. To start, I have selected which log groups I want to query. In this case, I select a log group of a Lambda function that I want to inspect and choose Run query.

In the Pattern tab, I see the patterns that have been found in these log groups. One of the patterns seems to be an error. I can select it to quickly add it as a filter to my query and focus on the logs that contain this pattern. For now, I choose the magnifying glass icon to analyze the pattern.

Console screenshot.

In the Pattern inspect window, a histogram with the occurrences of the pattern in the selected time period is shown. After the histogram, samples from the logs are provided.

Console screenshot.

The variable parts of the pattern (such as numbers) have been extracted as “tokens.” I select the Token values tab to see the values for a token. I can select a token value to quickly add it as a filter to the query and focus on the logs that contain this pattern with this specific value.

Console screenshot.

I can also look at the Related patterns tab to see other logs that typically occurred at the same time as the pattern I am analyzing. For example, if I am looking at an ERROR log that was always written alongside a DEBUG log showing more details, I would see that relationship there.

Comparing logs with a previous period
To better understand what is happening, I choose the Compare button in the time interval selector. This updates the query to compare results with a previous period. For example, I choose Previous day to see what changed compared to yesterday.

Console screenshot.

In the Patterns tab, I notice that there has actually been a 10 percent decrease in the number of errors, so the current situation might not be too bad.

I choose the magnifying glass icon on the pattern with severity type ERROR to see a full comparison of the two time periods. The graph overlaps the occurrences of the pattern over the two periods (now and yesterday in this case) inside the selected time range (one hour).

Console screenshot.

Errors are decreasing but are still there. To reduce those errors, I make some changes to the application. I come back after some time to compare the logs, and a new ERROR pattern is found that was not present in the previous time period.

Console screenshot.

My update probably broke something, so I roll back to the previous version of the application. For now, I’ll keep it as it is because the number of errors is acceptable for my use case.

Detecting anomalies in the log
I am reassured by the decrease in errors that I discovered comparing the logs. But how can I know if something unexpected is happening? Anomaly detection for CloudWatch Logs looks for unexpected patterns in the logs as they are processed during ingestion and can be enabled at log group level.

I select Log groups in the navigation pane and type a filter to see the same log group I was looking at before. I choose Configure in the Anomaly detection column and select an Evaluation frequency of 5 minutes. Optionally, I can use a longer interval (up to 60 minutes) and add patterns to process only specific log events for anomaly detection.

After I activate anomaly detection for this log group, incoming logs are constantly evaluated against historical baselines. I wait for a few minutes and, to see what has been found, I choose Log anomalies from the Logs section of the navigation pane.

Console screenshot.

To simplify this view, I can suppress anomalies that I am not interested in following. For now, I choose one of the anomalies in order to inspect the corresponding pattern in a way similar to before.

Console screenshot.

After this additional check, I am convinced there are no urgent issues with my application. With all the insights I collected with these new capabilities, I can now focus on the errors in the logs to understand how to solve them.

Things to know
Amazon CloudWatch automated log pattern analytics is available today in all commercial AWS Regions where Amazon CloudWatch Logs is offered excluding the China (Beijing), the China (Ningxia), and Israel (Tel Aviv) Regions.

The patterns and compare query features are charged according to existing Logs Insights query costs. Comparing a one-hour time period against another one-hour time period is equivalent to running a single query over a two-hour time period. Anomaly detection is included as part of your log ingestion fees, and there is no additional charge for this feature. For more information, see CloudWatch pricing.

Simplify how you analyze logs with CloudWatch automated log pattern analytics.

Danilo

New Amazon CloudWatch log class for infrequent access logs at a reduced price

Post Syndicated from Marcia Villalba original https://aws.amazon.com/blogs/aws/new-amazon-cloudwatch-log-class-for-infrequent-access-logs-at-a-reduced-price/

Amazon CloudWatch Logs announces today a new log class called Infrequent Access. This new log class offers a tailored set of capabilities at a lower cost for infrequently accessed logs, enabling customers to consolidate all their logs in one place in a cost-effective manner.

As customers’ applications continue to scale and grow, so does the volume of logs generated. To limit the increase of logging costs, many customers are forced to make hard trade-offs. For example, some customers limit the logs generated by their applications, which can hinder the visibility of the application, or choose a different solution for different log types, which adds complexity and inefficiencies in managing different logging solutions. For instance, customers may send logs needed for real-time analytics and alerting to CloudWatch Logs and send more detailed logs needed for debugging and troubleshooting to a lower-cost solution that doesn’t have as many features as CloudWatch. In the end, these workarounds can impact the observability of the application, because customers need to navigate across multiple solutions to see their logs.

The Infrequent Access log class allows you to build a holistic observability solution using CloudWatch by centralizing all your logs in one place to ingest, query, and store your logs in a cost-efficient way. Infrequent Access is 50 percent lower per GB ingestion price than Standard log class. It provides a tailored set of capabilities for customers that don’t need advanced features like Live Tail, metric extraction, alarming, or data protection that the Standard log class provides. With Infrequent Access, you can still get the benefits of fully managed ingestion, storage, and the ability to deep dive using CloudWatch Logs Insights.

The following table shows a side-by-side comparison of the features that the new Infrequent Access and the Standard log classes have.

Feature Infrequent Access log class Standard log class
Fully managed ingestion and storage Available Available
Cross-account Available Available
Encryption with KMS Available Available
Logs Insights Available Available
Subscription filters / Export to S3 Not available Available
GetLogEvents / FilterLogEvents Not available Available
Contributor, Container, and Lambda Insights Not available Available
Metric filter and alerting Not available Available
Data protection Not available Available
Embedded metric format (EMF) Not available Available
Live Tail Not available Available

When to use the new Infrequent Access log class
Use the Infrequent Access log class when you have a new workload that doesn’t require advanced features provided by the Standard log class. One important consideration is that when you create a log group with a specific log class, you cannot change that log group log class afterward.

The Infrequent Access log class is suitable for debug logs or web server logs because they are quite verbose and rarely require any of the advanced functionality that the Standard log class provides.

Another good workload for the Infrequent Access log class is an Internet of Things (IoT) fleet sending detailed logs that are only accessed for after the fact forensic analysis after the event. In addition, the Infrequent Access log class is a good choice for workloads where logs need to be stored for compliance because they will be queried infrequently.

Getting started
To get started using the new Infrequent Access log class, create a new log group in the CloudWatch Logs console and select the new Infrequent Access log class. You can create logs groups with the new Infrequent Access log class not only from the AWS Management Console but also from the AWS Command Line Interface (AWS CLI), AWS CloudFormation, AWS Cloud Development Kit (AWS CDK), and AWS SDKs.

Create log group

Once you have the new log group created, you can start using it in your workloads. For this example, I will configure a web application to send debug logs to this log group. After a while of the web application executes for a while, you can go back to the log group, where you see a new log stream.

View log group

When you select a log stream, you will be directed to CloudWatch Logs Insights.

Log insights

Using the same familiar CloudWatch Logs Insights experience you get with Standard Class, you can create queries and search those logs to find relevant information, and you can analyze all the logs quickly in one place.

Available now
The new Infrequent Access log class is now available in all AWS Regions except the China and GovCloud Regions. You can start using it and enjoy a more cost-effective way to collect, store, and analyze your logs in a fully managed experience.

To learn more about the new log class, you can check the CloudWatch Logs user guide dedicated page for the Infrequent Access log class.

Marcia

AWS Weekly Roundup – EC2 DL2q instances, PartyRock, Amplify’s 6th birthday, and more – November 20, 2023

Post Syndicated from Channy Yun original https://aws.amazon.com/blogs/aws/aws-weekly-roundup-ec2-dl2q-instances-partyrock-amplifys-6th-birthday-and-more-november-20-2023/

Last week I saw an astonishing 160+ new service launches. There were so many updates that we decided to publish a weekly roundup again. This continues the same innovative pace of the previous week as we are getting closer to AWS re:Invent 2023.

Our News Blog team is also finalizing new blog posts for re:Invent to introduce awesome launches with service teams for your reading pleasure. Jeff Barr shared The Road to AWS re:Invent 2023 to explain our blogging journey and process. Please stay tuned in the next week!

Last week’s launches
Here are some of the launches that caught my attention last week:

Amazon EC2 DL2q instances – New DL2q instances are powered by Qualcomm AI 100 Standard accelerators and are the first to feature Qualcomm’s AI technology in the public cloud. With eight Qualcomm AI 100 Standard accelerators and 128 GiB of total accelerator memory, you can run popular generative artificial intelligence (AI) applications and extend to edge devices across smartphones, autonomous driving, personal compute, and extended reality headsets to develop and validate these AI workloads before deploying.

PartyRock for Amazon Bedrock – We introduced PartyRock, a fun and intuitive hands-on, generative AI app-building playground powered by Amazon Bedrock. You can experiment, learn all about prompt engineering, build mini-apps, and share them with your friends—all without writing any code or creating an AWS account.

You also can now access the Meta Llama 2 Chat 13B foundation model and Cohere Command Light, Embed English, and multilingual models for Amazon Bedrock.

AWS Amplify celebrates its sixth birthday – We announced six new launches; a new documentation site, support for Next.js 14 with our hosting and JavaScript library, added custom token providers and an automatic React Native social sign-in update to Amplify Auth, new ChangePassword and DeleteUser account settings components, and updated all Amplify UI packages to use new Amplify JavaScript v6. You can also use wildcard subdomains when using a custom domain with your Amplify application deployed to AWS Amplify Hosting.

Amplify docs site UI

Also check out other News Blog posts about major launches published in the past week:

Other AWS service launches
Here are some other bundled feature launches per AWS service:

Amazon Athena  – You can use a new cost-based optimizer (CBO) to enhance query performance based on table and column statistics, collected by AWS Glue Data Catalog and Athena JDBC 3.x driver, a new alternative that supports almost all authentication plugins. You can also use Amazon EMR Studio to develop and run interactive queries on Amazon Athena.

Amazon CloudWatch – You can use a new CloudWatch metric called EBS Stalled I/O Check to monitor the health of your Amazon EBS volumes, the regular expression for Amazon CloudWatch Logs Live Tail filter pattern syntax to search and match relevant log events, observability of SAP Sybase ASE database in CloudWatch Application Insights, and up to two stats commands in a Log Insights query to perform aggregations on the results.

Amazon CodeCatalyst – You can connect to a Amazon Virtual Private Cloud (Amazon VPC) from CodeCatalyst Workflows, provision infrastructure using Terraform within CodeCatalyst Workflows, access CodeCatalyst with your workforce identities configured in IAM Identity Center, and create teams made up of members of the CodeCatalyst space.

Amazon Connect – You can use a pre-built queue performance dashboard and Contact Lens conversational analytics dashboard to view and compare real-time and historical aggregated queue performance. You can use quick responses for chats, previously written formats such as typing in ‘/#greet’ to insert a personalized response, and scanning attachments to detect malware or other unwanted content.

AWS Glue – AWS Glue for Apache Spark added new six database connectors: Teradata, SAP HANA, Azure SQL, Azure Cosmos DB, Vertica, and MongoDB, as well as the native connectivity to Amazon OpenSearch Service.

AWS Lambda – You can see single pane view of metrics, logs, and traces in the AWS Lambda console and advanced logging controls to natively capture logs in JSON structured format. You can view the SAM template on the Lambda console and export the function’s configuration to AWS Application Composer. AWS Lambda also supports Java 21 and NodeJS 20 versions built on the new Amazon Linux 2023 runtime.

AWS Local Zones in Dallas – You can enable the new Local Zone in Dallas, Texas, us-east-1-dfw-2a, with Amazon EC2 C6i, M6i, R6i, C6gn, and M6g instances and Amazon EBS volume types gp2, gp3, io1, sc1, and st1. You can also access Amazon ECS, Amazon EKS, Application Load Balancer, and AWS Direct Connect in this new Local Zone to support a broad set of workloads at the edge.

Amazon Managed Streaming for Apache Kafka (Amazon MSK) – You can standardize access control to Kafka resources using AWS Identity and Access Management (IAM) and build Kafka clients for Amazon MSK Serverless written in all programming languages. These are open source client helper libraries and code samples for popular languages, including Java, Python, Go, and JavaScript. Also, Amazon MSK now supports an enhanced version of Apache Kafka 3.6.0 that offers generally available Tiered Storage and automatically sends you storage capacity alerts when you are at risk of exhausting your storage.

Amazon OpenSearch Service Ingestion – You can migrate your data from Elasticsearch version 7.x clusters to the latest versions of Amazon OpenSearch Service and use persistent buffering to protect the durability of incoming data.

Amazon RDS –Amazon RDS for MySQL now supports creating active-active clusters using the Group Replication plugin, upgrading MySQL 5.7 snapshots to MySQL 8.0, and Innovation Release version of MySQL 8.1.

Amazon RDS Custom for SQL Server extends point-in-time recovery support for up to 1,000 databases, supports Service Master Key Retention to use transparent data encryption (TDE), table- and column-level encryption, DBMail and linked servers, and use SQL Server Developer edition with the bring your own media (BYOM).

Additionally, Amazon RDS Multi-AZ deployments with two readable standbys now supports minor version upgrades and system maintenance updates with typically less than one second of downtime when using Amazon RDS Proxy.

AWS Partner Central – You can use an improved user experience in AWS Partner Central to build and promote your offerings and the new Investments tab in the Partner Analytics Dashboard to gain actionable insights. You can now link accounts and associated users between Partner Central and AWS Marketplace and use an enhanced co-sell experience with APN Customer Engagements (ACE) manager.

Amazon QuickSight – You can programmatically manage user access and custom permissions support for roles to restrict QuickSight functionality to the QuickSight account for IAM Identity Center and Active Directory using APIs. You can also use shared restricted folders, a Contributor role and support for data source asset types in folders and the Custom Week Start feature, an addition designed to enhance the data analysis experience for customers across diverse industries and social contexts.

AWS Trusted Advisor – You can use new APIs to programmatically access Trusted Advisor best practices checks, recommendations, and prioritized recommendations and 37 new Amazon RDS checks that provide best practices guidance by analyzing DB instance configuration, usage, and performance data.

There’s a lot more launch news that I haven’t covered. See AWS What’s New for more details.

See you virtually in AWS re:Invent
AWS re:Invent 2023Next week we’ll hear the latest from AWS, learn from experts, and connect with the global cloud community in Las Vegas. If you come, check out the agenda, session catalog, and attendee guides before your departure.

If you’re not able to attend re:Invent in person this year, we’re offering the option to livestream our Keynotes and Innovation Talks. With the registration for online pass, you will have access to on-demand keynote, Innovation Talks, and selected breakout sessions after the event.

Channy

Introducing advanced logging controls for AWS Lambda functions

Post Syndicated from David Boyne original https://aws.amazon.com/blogs/compute/introducing-advanced-logging-controls-for-aws-lambda-functions/

This post is written by Nati Goldberg, Senior Solutions Architect and Shridhar Pandey, Senior Product Manager, AWS Lambda

Today, AWS is launching advanced logging controls for AWS Lambda, giving developers and operators greater control over how function logs are captured, processed, and consumed.

This launch introduces three new capabilities to provide a simplified and enhanced default logging experience on Lambda.

First, you can capture Lambda function logs in JSON structured format without having to use your own logging libraries. JSON structured logs make it easier to search, filter, and analyze large volumes of log entries.

Second, you can control the log level granularity of Lambda function logs without making any code changes, enabling more effective debugging and troubleshooting.

Third, you can also set which Amazon CloudWatch log group Lambda sends logs to, making it easier to aggregate and manage logs at scale.

Overview

Being able to identify and filter relevant log messages is essential to troubleshoot and fix critical issues. To help developers and operators monitor and troubleshoot failures, the Lambda service automatically captures and sends logs to CloudWatch Logs.

Previously, Lambda emitted logs in plaintext format, also known as unstructured log format. This unstructured format could make the logs challenging to query or filter. For example, you had to search and correlate logs manually using well-known string identifiers such as “START”, “END”, “REPORT” or the request id of the function invocation. Without a native way to enrich application logs, you needed custom work to extract data from logs for automated analysis or to build analytics dashboards.

Previously, operators could not control the level of log detail generated by functions. They relied on application development teams to make code changes to emit logs with the required granularity level, such as INFO, DEBUG, or ERROR.

Lambda-based applications often comprise microservices, where a single microservice is composed of multiple single-purpose Lambda functions. Before this launch, Lambda sent logs to a default CloudWatch log group created with the Lambda function with no option to select a log group. Now you can aggregate logs from multiple functions in one place so you can uniformly apply security, governance, and retention policies to your logs.

Capturing Lambda logs in JSON structured format

Lambda now natively supports capturing structured logs in JSON format as a series of key-value pairs, making it easier to search and filter logs more easily.

JSON also enables you to add custom tags and contextual information to logs, enabling automated analysis of large volumes of logs to help understand the function performance. The format adheres to the OpenTelemetry (OTel) Logs Data Model, a popular open-source logging standard, enabling you to use open-source tools to monitor functions.

To set the log format in the Lambda console, select the Configuration tab, choose Monitoring and operations tools on the left pane, then change the log format property:

Currently, Lambda natively supports capturing application logs (logs generated by the function code) and system logs (logs generated by the Lambda service) in JSON structured format.

This is for functions that use non-deprecated versions of Python, Node.js, and Java Lambda managed runtimes, when using Lambda recommended logging methods such as using logging library for Python, console object for Node.js, and LambdaLogger or Log4j for Java.

For other managed runtimes, Lambda currently only supports capturing system logs in JSON structured format. However, you can still capture application logs in JSON structured format for these runtimes by manually configuring logging libraries. See configuring advanced logging controls section in the Lambda Developer Guide to learn more. You can also use Powertools for AWS Lambda to capture logs in JSON structured format.

Changing log format from text to JSON can be a breaking change if you parse logs in a telemetry pipeline. AWS recommends testing any existing telemetry pipelines after switching log format to JSON.

Working with JSON structured format for Node.js Lambda functions

You can use JSON structured format with CloudWatch Embedded Metric Format (EMF) to embed custom metrics alongside JSON structured log messages, and CloudWatch automatically extracts the custom metrics for visualization and alarming. However, to use JSON log format along with EMF libraries for Node.js Lambda functions, you must use the latest version of the EMF client library for Node.js or the latest version of Powertools for AWS Lambda (TypeScript) library.

Configuring log level granularity for Lambda function

You can now filter Lambda logs by log level, such as ERROR, DEBUG, or INFO, without code changes. Simplified log level filtering enables you to choose the required logging granularity level for Lambda functions, without sifting through large volumes of logs to debug errors.

You can specify separate log level filters for application logs (which are logs generated by the function code) and system logs (which are logs generated by the Lambda service, such as START and REPORT log messages). Note that log level controls are only available if the log format of the function is set to JSON.

The Lambda console allows setting both the Application log level and System log level properties:

You can define the granularity level of each log event in your function code. The following statement prints out the event input of the function, emitted as a DEBUG log message:

console.debug(event);

Once configured, log events emitted with a lower log level than the one selected are not published to the function’s CloudWatch log stream. For example, setting the function’s log level to INFO results in DEBUG log events being ignored.

This capability allows you to choose the appropriate amount of logs emitted by functions. For example, you can set a higher log level to improve the signal-to-noise ratio in production logs, or set a lower log level to capture detailed log events for testing or troubleshooting purposes.

Customizing Lambda function’s CloudWatch log group

Previously, you could not specify a custom CloudWatch log group for functions, so you could not stream logs from multiple functions into a shared log group. Also, to set custom retention policy for multiple log groups, you had to create each log group separately using a pre-defined name (for example, /aws/lambda/<function name>).

Now you can select a custom CloudWatch log group to aggregate logs from multiple functions automatically within an application in one place. You can apply security, governance, and retention policies at the application level instead of individually to every function.

To distinguish between logs from different functions in a shared log group, each log stream contains the Lambda function name and version.

You can share the same log group between multiple functions to aggregate logs together. The function’s IAM policy must include the logs:CreateLogStream and logs:PutLogEvents permissions for Lambda to create logs in the specified log group. The Lambda service can optionally create these permissions, when you configure functions in the Lambda console.

You can set the custom log group in the Lambda console by entering the destination log group name. If the entered log group does not exist, Lambda creates it automatically.

Advanced logging controls for Lambda can be configured using Lambda APIAWS Management ConsoleAWS Command Line Interface (CLI), and infrastructure as code (IaC) tools such as AWS Serverless Application Model (AWS SAM) and AWS CloudFormation.

Example of Lambda advanced logging controls

This section demonstrates how to use the new advanced logging controls for Lambda using AWS SAM to build and deploy the resources in your AWS account.

Overview

The following diagram shows Lambda functions processing newly created objects inside an Amazon S3 bucket, where both functions emit logs into the same CloudWatch log group:

The architecture includes the following steps:

  1. A new object is created inside an S3 bucket.
  2. S3 publishes an event using S3 Event Notifications to Amazon EventBridge.
  3. EventBridge triggers two Lambda functions asynchronously.
  4. Each function processes the object to extract labels and text, using Amazon Rekognition and Amazon Textract.
  5. Both functions then emit logs into the same CloudWatch log group.

This uses AWS SAM to define the Lambda functions and configure the required logging controls. The IAM policy allows the function to create a log stream and emit logs to the selected log group:

DetectLabelsFunction:
    Type: AWS::Serverless::Function 
    Properties:
      CodeUri: detect-labels/
      Handler: app.lambdaHandler
      Runtime: nodejs18.x
      Policies:
        ...
        - Version: 2012-10-17
          Statement:
            - Sid: CloudWatchLogGroup
              Action: 
                - logs:CreateLogStream
                - logs:PutLogEvents
              Resource: !GetAtt CloudWatchLogGroup.Arn
              Effect: Allow
      LoggingConfig:
        LogFormat: JSON 
        ApplicationLogLevel: DEBUG 
        SystemLogLevel: INFO 
        LogGroup: !Ref CloudWatchLogGroup 

Deploying the example

To deploy the example:

  1. Clone the GitHub repository and explore the application.
    git clone https://github.com/aws-samples/advanced-logging-controls-lambda/
    
    cd advanced-logging-controls-lambda
  2. Use AWS SAM to build and deploy the resources to your AWS account. This compiles and builds the application using npm, and then populate the template required to deploy the resources:
    sam build
  3. Deploy the solution to your AWS account with a guided deployment, using AWS SAM CLI interactive flow:
    sam deploy --guided
  4. Enter the following values:
    • Stack Name: advanced-logging-controls-lambda
    • Region: your preferred Region (for example, us-east-1)
    • Parameter UploadsBucketName: enter a unique bucket name.
    • Accept the rest of the initial defaults.
  5. To test the application, use the AWS CLI to copy the sample image into the S3 bucket you created.
    aws s3 cp samples/skateboard.jpg s3://example-s3-images-bucket

Explore CloudWatch Logs to view the logs emitted into the log group created, AggregatedLabelsLogGroup:

The DetectLabels Lambda function emits DEBUG log events in JSON format to the log stream. Log events with the same log level from the ExtractText Lambda function are omitted. This is a result of the different application log level settings for each function (DEBUG and INFO).

You can also use CloudWatch Logs Insights to search, filter, and analyze the logs in JSON format using this sample query:

You can see the results:

Conclusion

Advanced logging controls for Lambda give you greater control over logging. Use advanced logging controls to control your Lambda function’s log level and format, allowing you to search, query, and filter logs to troubleshoot issues more effectively.

You can also choose the CloudWatch log group where Lambda sends your logs. This enables you to aggregate logs from multiple functions into a single log group, apply retention, security, governance policies, and easily manage logs at scale.

To get started, specify the required settings in the Logging Configuration for any new or existing Lambda functions.

Advanced logging controls for Lambda are available in all AWS Regions where Lambda is available at no additional cost. Learn more about AWS Lambda Advanced Logging Controls.

For more serverless learning resources, visit Serverless Land.

Managing AWS Lambda runtime upgrades

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

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

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

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

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

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

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

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

Specifying Lambda runtimes

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

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

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

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

Why does Lambda deprecate runtimes?

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

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

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

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

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

Can Lambda automatically upgrade my runtime?

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

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

Lambda function versions

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

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

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

Version 1 using Node.js 14.x

Version 1 using Node.js 14.x

Version 2 using Node.js 18.x

Version 2 using Node.js 18.x

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

Managing function runtime upgrades

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

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

Identifying impacted functions

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

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

AWS Trusted Advisor Lambda functions using deprecated runtimes

AWS Trusted Advisor Lambda functions using deprecated runtimes

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

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

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

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

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

Dashboard showing AWS Config data

Dashboard showing AWS Config data

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

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

Track invocations in Amazon CloudWatch metrics

Track invocations in Amazon CloudWatch metrics

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

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

Limiting runtime usage

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

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

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

Upgrade best practices

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

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

Using runtimes after deprecation

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

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

Conclusion

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

For more serverless learning resources, visit Serverless Land.

Enhanced Amazon CloudWatch metrics for Amazon EventBridge

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

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

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

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

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

What’s new?

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

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

EventBridge now has the following metrics:

Metric Description Dimensions and Units
PutEventsLatency The time taken per PutEvents API operation

None

Units: Milliseconds

PutEventsRequestSize The size of the PutEvents API request in bytes

None

Units: Bytes

MatchedEvents Number of events that matched with any rule, or a specific rule None
RuleName,
EventBusName,
EventSourceName

Units: Count

ThrottledRules The number of times rule execution was throttled.

None, RuleName

Unit: Count

PutEventsApproximateCallCount Approximate total number of calls in PutEvents API calls.

None

Units: Count

PutEventsApproximateThrottledCount Approximate number of throttled requests in PutEvents API calls.

None

Units: Count

PutEventsApproximateFailedCount Approximate number of failed PutEvents API calls.

None

Units: Count

PutEventsApproximateSuccessCount Approximate number of successful PutEvents API calls.

None

Units: Count

PutEventsEntriesCount The number of event entries contained in a PutEvents request.

None

Units: Count

PutEventsFailedEntriesCount The number of event entries contained in a PutEvents request that failed to be ingested.

None

Units: Count

PutPartnerEventsApproximateCallCount Approximate total number of calls in PutPartnerEvents API calls. (visible in Partner’s account)

None

Units: Count

PutPartnerEventsApproximateThrottledCount Approximate number of throttled requests in PutPartnerEvents API calls. (visible in Partner’s account)

None

Units: Count

PutPartnerEventsApproximateFailedCount Approximate number of failed PutPartnerEvents API calls. (visible in Partner’s account)

None

Units: Count

PutPartnerEventsApproximateSuccessCount Approximate number of successful PutPartnerEvents API calls. (visible in Partner’s account)

None

Units: Count

PutPartnerEventsEntriesCount The number of event entries contained in a PutPartnerEvents request.

None

Units: Count

PutPartnerEventsFailedEntriesCount The number of event entries contained in a PutPartnerEvents request that failed to be ingested.

None

Units: Count

PutPartnerEventsLatency The time taken per PutPartnerEvents API operation (visible in Partner’s account)

None

Units: Milliseconds

InvocationsCreated Number of times a target is invoked by a rule in response to an event. One invocation attempt represents a single count for this metric.

None

Units: Count

InvocationAttempts Number of times EventBridge attempted invoking a target.

None

Units: Count

SuccessfulInvocationAttempts Number of times target was successfully invoked.

None

Units: Count

RetryInvocationAttempts The number of times a target invocation has been retried.

None

Units: Count

IngestiontoInvocationStartLatency The time to process events, measured from when an event is ingested by EventBridge to the first invocation of a target. None,
RuleName,
EventBusName

Units: Milliseconds

IngestiontoInvocationCompleteLatency The time taken from event Ingestion to completion of the first successful invocation attempt None,
RuleName,
EventBusName

Units: Milliseconds

Use-cases for these metrics

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

Monitor service usage based on Service Quotas

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

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

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

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

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

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

Invocations throttle limit in transactions per second

Enhanced observability

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

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

Enhanced observability

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

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

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

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

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

Failover events to secondary Region

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

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

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

Conclusion

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

For more serverless learning resources, visit Serverless Land.

Journey to Cloud-Native Architecture Series #7:  Using Containers and Cell-based design for higher resiliency and efficiency

Post Syndicated from Anuj Gupta original https://aws.amazon.com/blogs/architecture/journey-to-cloud-native-architecture-series-7-using-containers-and-cell-based-design-for-higher-resiliency-and-efficiency/

In our previous Journey to Cloud-Native blogposts, we talked about evolving our architecture to become more scalable, secure, and cost effective to handle hyperscale requirements. In this post, we take these next steps: 1/ containerizing our applications to improve resource efficiency, and, 2/ using cell-based design to improve resiliency and time to production.

Containerize applications for standardization and scale

Standardize container orchestration tooling

Selecting the right service for your use case requires considering your organizational needs and skill sets for managing container orchestrators at scale. Our team chose Amazon Elastic Kubernetes Service (EKS), because we have the skills and experience working with Kubernetes. In addition, our leadership was committed to open source, and the topology awareness feature aligned with our resiliency requirements.

For containerizing our Java and .NET applications, we used AWS App2Container (A2C) to create deployment artifacts. AWS A2C reduced the time needed for dependency analysis, and created artifacts that we could plug into our deployment pipeline. A2C supports EKS Blueprints with GitOps which helps reduce the time-to-container adoption. Blueprints also improved consistency and follows security best practices. If you are unsure on the right tooling for running containerized workloads, you can refer to Choosing an AWS container service.

Identify the right tools for logging and monitoring

Logging in to the container environment adds some complexity due to the dynamic number of short-lived log sources. For proper tracing of events, we needed a way to collect logs from all of the system components and applications. We set up the Fluent Bit plugin to collect logs and send them to Amazon CloudWatch.

We used CloudWatch Container Insights for Prometheus for scraping Prometheus metrics from containerized applications. It allowed us to use existing tools (Amazon CloudWatch and Prometheus) to build purpose-built dashboards for different teams and applications. We created dashboards in Amazon Managed Grafana by using the native integration with Amazon CloudWatch. These tools took away the heavy lifting of managing logging and container monitoring from our teams.

Managing resource utilization

In hyperscale environments, a noisy neighbor container can consume all the resources for an entire cluster. Amazon EKS provides the ability to define and apply requests and limits for pods and containers. These values determine the minimum and maximum amount of a resource that a container can have. We also used resource quotas to configure the total amount of memory and CPU that can be used by all pods running in a namespace.

In order to better understand the resource utilization and cost of running individual applications and pods, we implemented Kubecost, using the guidance from Multi-cluster cost monitoring for Amazon EKS using Kubecost and Amazon Managed Service for Prometheus.

Scaling cluster and applications dynamically

With Amazon EKS, the Kubernetes metric server GitHub collects resource metrics from Kubelets and exposes them in the Kubernetes API server. This is accomplished through the metrics API for use by the Horizontal Pod Autoscaler (HPA) and Vertical Pod Autoscaler (VPA). The HPA consumes these metrics to provide horizontal scaling by increasing the number of replicas to distribute your workloads. The VPA uses these metrics to dynamically adjust for pod resources like CPU/Memory reservations.

With metrics server integration, Amazon EKS takes away the operational complexity of scaling applications and provides more granular controls on scaling applications dynamically. We recommend that hyperscale customers consider HPA as their preferred application autoscaler because it provides resiliency (increased number of replicas) in addition to scaling. On the cluster level, Karpenter provides rapid provisioning and de-provisioning of large numbers of diverse compute resources, in addition to providing cost effective usage. It helps applications hyperscale to meet business growth needs.

Build rollback strategies for failure management

In hyperscale environments, deployment rollouts typically have a low margin for errors and require close to zero downtime. We implemented progressive rollouts, canary deployments, and automated rollbacks to reduce risk for production deployments. We used key performance indicators (KPIs) like application response time and error rates to determine whether to continue or to rollback our deployment. We leveraged the integration with Prometheus to collect metrics and measure KPIs.

Improving resilience and scale using cell-based design

We still needed to handle black swan events and minimize the impact of unexpected failures or scaling events to make them more resilient. We came up with a design that creates independently deployable units of applications with contained fault isolation boundaries. The new design uses a cell-based approach, along with a shuffle sharding technique to further improve resiliency. Peter Vosshall’s re:Invent talk details this approach.

First, we created a cell-based architecture as described in Reducing the Scope of Impact with Cell-Based Architecture. Second, we applied shuffle sharding as described in What about shuffle-sharding?, to further control system impact in case of black swan events.

In Figure 1 we see two components:

  • A light-weight routing layer that manages the traffic routing of incoming requests to the cells.
  • The cells themselves, which are independent units with isolated boundaries to prevent system wide impact.
High level cell-based architecture

Figure 1. High level cell-based architecture

Figure 1. High Level cell-based architecture

We used best practices from Guidance for Cell-based Architecture on AWS for implementation of our cell-based architectures and routing layer. In order to minimize the risk of failure in the routing layer, we made it the thinnest layer and tested robustness for all possible scenarios. We used a hash function to map cells to customers and stored the mapping in a highly scaled and resilient data store, Amazon DynamoDB. This layer eases the addition of new cells, and provides for a horizontal-scaled application environment to gracefully handle the hypergrowth of customers or orders.

In Figure 2, we revisit the architecture as mentioned on Blog #3 of our series. To manage AWS service limits and reduce blast radius, we spread our applications into multiple AWS accounts.

Architecture from Blog #3

Figure 2. Architecture from Blog #3

Here’s how the new design looks when implemented on AWS:

Cell-based architecture

Figure 3a. Cell-based architecture

Figure 3a uses Amazon EKS instead of Amazon EC2 for deploying applications, and uses a light-weight routing layer for incoming traffic. An Amazon Route 53 configuration was deployed in a networking shared services AWS account. We have multiple isolated boundaries in our design — an AWS availability zone, an AWS Region, and an AWS account. This makes the architecture more resilient and flexible in hypergrowth.

We used our AWS availability zones as a cell boundary. We implemented shuffle sharding to map customers to multiple cells to reduce impact in case one of the cells goes down. We also used shuffle sharding to scale the number of cells that can process a customer request, in case we have unprecedented requests from a customer.

Cell-based design to handle Black swan events Black swan event mitigation

Let’s discuss how our cell-based application will react to black swan events. First, we need to align the pods into cell groups in the Amazon EKS cluster. Now we can observe how each deployment is isolated from the other deployments. Only the routing layer, which include Amazon Route 53, Amazon DynamoDB, and the load balancer, is common across the system.

Cell-based architecture zoomed in on EKS

Figure 3b. Cell-based architecture zoomed in on EKS

Without a cell-based design, a black swan event could take down our application in the initial attack. Now that our application is spread over three cells, the worst case for an initial attack is 33% of our application’s capacity. We now have resiliency boundaries at the node level and availability zones.

Conclusion

In this blog post, we discussed how containerizing applications can improve resource efficiency and help standardize tooling and processes. This reduces engineering overhead of scaling applications. We talked about how adopting cell-based design and shuffle sharding further improves the resilience posture of our applications, our ability to manage failures, and handle unexpected large scaling events.

Further reading: