AWS Bedrock Monitoring | Datadog

AWS Bedrock Monitoring

Monitor Amazon Bedrock metrics and logs with Datadog. Optimize your LLM model usage, determine how users are querying your AI applications, and reduce pain points.

bedrock header

Amazon Bedrock is a fully managed service that offers foundation models (FMs) and other tools for building generative AI applications. Customers can fine-tune their FMs to invoke tasks such as text generation, content creation, and chatbot Q&A — without provisioning or managing any infrastructure.

Datadog’s latest integration with Amazon Bedrock allows you to monitor your Foundation Model usage, track API performance, and document your error rate with runtime metrics and logs. Once installed, Datadog will automatically collect Bedrock metrics and logs. Collecting Bedrock metrics allows you to monitor how frequently your models are used, and discover opportunities to optimize model usage. Datadog can also monitor Bedrock API calls via CloudWatch logs, giving clarity into how customers are using your LLM, and how it responds.

 

Monitor Your AI Models' API Performance and Errors

  • Easy installation: After installing the integration, Datadog will immediately begin collecting runtime metrics and logs.
  • Visualize telemetry data:Amazon Bedrock data is displayed in an out-of-the-box Datadog dashboard, giving insight into your Foundation Models' performance.
  • Monitor latency: Track each model's performance and determine which model could be slowing down your AI application.
bedrock1

Optimize Your Model Usage

  • Discover how users are using your AI: Track how frequently your models are invoked by customers and discover more about how they use your AI applications.
  • Monitor input and output token count: Gauge the complexity of the inputs and outputs your model processes.
  • Optimize your models: Discover which models are most heavily used, and allocate more throughput accordingly to get faster processing times.
bedrock2

Gain Insights into Model Prompts and Responses with Logs

  • Monitor API Calls: Monitor Bedrock API calls through CloudWatch logs, displayed in the Datadog dashboard.
  • Gain insights into user queries: Analyze logs to analyze how users are querying your models, and identify common trends and pain points.
  • Reduce frustration: Discover whether users are using your models as intended, and reduce pain points that product teams may have missed.
bedrock3

The Essential Monitoring and Security Platform for the Cloud Age

Datadog brings together end-to-end traces, metrics, and logs to make your applications, infrastructure, and third-party services entirely observable.

Platform Diagram

Next-generation AI Monitoring

Monitor and your entire AI and machine learning stack with Datadog.

watchdog-apm-illustration.png

Amazon Bedrock

Monitor your FM usage, API performance, and error rate with runtime metrics and logs.

tracesearch-apm-illustrationv2.png

OpenAI

Monitor token consumption, API performance, and more.

servicemap-apm-illustration.png

NVIDIA DCGM Exporter

Gather metrics from NVIDIA’s discrete GPUs, essential to parallel computing.

Loved & Trusted by Thousands

Washington Post logo 21st Century Fox Home Entertainment logo Peloton logo Samsung logo Comcast logo Nginx logo

AI Monitoring Resources

Learn about how Datadog can help you monitor your entire AI stack, from Amazon Bedrock to LLM models.

Datadog AI Monitoring Starter Kit