Instrument Your AWS Lambda Fleet With Remote Bulk Instrumentation | Datadog

Instrument your AWS Lambda fleet with remote bulk instrumentation

Author Sumedha Mehta

Published: June 26, 2024

In today’s rapidly evolving cloud environments, it’s more important than ever to have end-to-end visibility into transaction-level data for your production serverless applications to be able to solve issues quickly and make your services efficient. However, ensuring that all critical serverless functions are adequately instrumented for performance and security monitoring can be a time-consuming task, especially during critical incidents. The decentralized nature of microservice development often leads to sprawl in instrumentation best practices, fragmented internal documentation, and varied toolsets. And the dynamic nature of serverless functions can make it difficult to maintain continuous instrumentation. This can lead to incomplete observability and instability or blindspots within applications when issues or incidents arise.

To help address these challenges, Datadog supports remote bulk instrumentation of your Lambda functions, now in Preview. This enables you to add features like enhanced metrics and distributed tracing to any number of Lambda functions without requiring changes to your CI/CD pipeline or infrastructure-as-code (IaC) configuration.

Instrument your Lambda functions directly from the Datadog App

Maintaining full monitoring coverage of your Lambda functions can often require chasing down individual application teams to instrument serverless apps, or overhauling IaC or code pipelines to make sure that functions stay instrumented after any modifications and redeployments.

Datadog’s remote bulk instrumentation for Lambda provides a seamless configuration option for organizations that need to set and forget instrumentation across their fleet of AWS Lambda functions. After you enable the AWS integration, Datadog collects information about your AWS account and provides a list of all detected Lambda functions while indicating which have been instrumented. Minutes after enabling bulk instrumentation, the Datadog Lambda Instrumenter instruments the selected functions with the Datadog Lambda extension and ensures that they stay instrumented regardless of any further deployments of the Lambda function.

With remote bulk instrumentation for Lambda, your teams can reduce the amount of time that they spend modifying deployment pipelines or infrastructure configurations. Adding serverless instrumentation on-the-fly means organizations can also get visibility into execution-level Lambda traces during critical incidents, even when serverless apps may not have initially been instrumented.

Granularly control instrumentation

Remote bulk instrumentation for Lambda is scoped to a specific AWS account ID and Region. After choosing the account to start bulk instrumenting, you can identify via tags or function names which functions you want to enable bulk instrumentation for. You can also select via function name which ones you want to exclude. Once rules are set, the Instrumenter ensures that your Lambda functions stay instrumented, even after subsequent deployments.

Select which Lambda functions within your AWS account and Region you want to instrument.

Once you’ve set up the rules for your Lambda functions, you can deploy the AWS CloudFormation stack with a Datadog API key to your selected AWS account and Region. The CloudFormation template gives you the ability to select configuration options such as pinning the Lambda extension version and ensuring that your instrumented functions have a minimum memory verison.

Deploy the AWS CloudFormation stack to configure instrumented functions.

Simplify serverless observability

Datadog’s remote bulk instrumentation for Lambda streamlines serverless application management by automatically ensuring your AWS Lambda functions are instrumented without needing changes to IaC or code pipelines. This feature reduces the time teams spend on instrumentation, supports real-time tracing during critical incidents, and enables central operations teams to maintain oversight without coordinating with individual app owners. Once set, functions remain instrumented with the Datadog Lambda Extension through subsequent deployments, ensuring continuous and reliable monitoring.

This feature is currently in Preview. Request access here. If you’re not a customer, sign up for a 14-day .