Amazon Bedrock AgentCore Evaluations is now generally available, providing automated quality assessment for AI agents. Evaluations enables developers to monitor agent quality through continuous evaluation of production traffic, validate changes through testing workflows, and measure agent performance against defined expectations. AgentCore Evaluations offers two evaluation types. Online evaluation continuously monitors agent performance in production by sampling and scoring live traces. On-demand evaluation enables teams to test agents programmatically, supporting regression testing in CI/CD pipelines and interactive development workflows.

Teams can evaluate agents using 13 built-in evaluators for response quality, safety, task completion, and tool usage. Developers can also use Ground Truth to measure agent performance against expectations, including reference answers for response validation, behavioral assertions for session-level goals, and expected tool execution sequences. For domain-specific requirements, teams can configure custom evaluators using their choice of prompts and model for LLM-based evaluation, or implement custom logic in Python or JavaScript through Lambda-hosted functions for code-based evaluation. Evaluations integrates with AgentCore Observability for unified monitoring and real-time alerts.

AgentCore Evaluations is available in nine AWS Regions: US East (N. Virginia), US East (Ohio), US West (Oregon), Asia Pacific (Mumbai), Asia Pacific (Singapore), Asia Pacific (Sydney), Asia Pacific (Tokyo), Europe (Frankfurt), Europe (Ireland).
Learn more about Amazon Bedrock AgentCore Evaluations through the documentation, and get started with the AgentCore Starter Toolkit

 

​Amazon Bedrock AgentCore Evaluations is now generally available, providing automated quality assessment for AI agents. Evaluations enables developers to monitor agent quality through continuous evaluation of production traffic, validate changes through testing workflows, and measure agent performance against defined expectations. AgentCore Evaluations offers two evaluation types. Online evaluation continuously monitors agent performance in production by sampling and scoring live traces. On-demand evaluation enables teams to test agents programmatically, supporting regression testing in CI/CD pipelines and interactive development workflows. Teams can evaluate agents using 13 built-in evaluators for response quality, safety, task completion, and tool usage. Developers can also use Ground Truth to measure agent performance against expectations, including reference answers for response validation, behavioral assertions for session-level goals, and expected tool execution sequences. For domain-specific requirements, teams can configure custom evaluators using their choice of prompts and model for LLM-based evaluation, or implement custom logic in Python or JavaScript through Lambda-hosted functions for code-based evaluation. Evaluations integrates with AgentCore Observability for unified monitoring and real-time alerts. AgentCore Evaluations is available in nine AWS Regions: US East (N. Virginia), US East (Ohio), US West (Oregon), Asia Pacific (Mumbai), Asia Pacific (Singapore), Asia Pacific (Sydney), Asia Pacific (Tokyo), Europe (Frankfurt), Europe (Ireland). Learn more about Amazon Bedrock AgentCore Evaluations through the documentation, and get started with the AgentCore Starter Toolkit  

Today, AWS Deadline Cloud introduces three powerful new fleet scaling options that give you greater flexibility in managing your render farm capacity and performance: worker idle duration, standby worker count, and scale out rate. AWS Deadline Cloud is a fully managed service that helps creative teams efficiently manage and scale their rendering workloads in the cloud.

These new options give you direct control over balancing rendering speed and efficiency. Configurable worker idle duration allows you to specify how long workers remain available after completing a job, eliminating wait times between job submissions and speeding up artist’s iteration workflow. Standby worker count maintains a pool of pre-warmed, idle workers that are immediately available at job submission so your renders start right away. Scale out rate lets you configure how quickly your fleet scales, up to 500 workers per minute, giving you the control you need to match your infrastructure needs.

These flexible scaling controls are now available in AWS Deadline Cloud. To learn more, visit the AWS Deadline Cloud documentation.

 

​Today, AWS Deadline Cloud introduces three powerful new fleet scaling options that give you greater flexibility in managing your render farm capacity and performance: worker idle duration, standby worker count, and scale out rate. AWS Deadline Cloud is a fully managed service that helps creative teams efficiently manage and scale their rendering workloads in the cloud. These new options give you direct control over balancing rendering speed and efficiency. Configurable worker idle duration allows you to specify how long workers remain available after completing a job, eliminating wait times between job submissions and speeding up artist’s iteration workflow. Standby worker count maintains a pool of pre-warmed, idle workers that are immediately available at job submission so your renders start right away. Scale out rate lets you configure how quickly your fleet scales, up to 500 workers per minute, giving you the control you need to match your infrastructure needs.
These flexible scaling controls are now available in AWS Deadline Cloud. To learn more, visit the AWS Deadline Cloud documentation.  

Today we are announcing the release of Aurora DSQL connectors for .NET (Npgsql) and Rust (SQLx) that make it easy to build .NET and Rust applications on Aurora DSQL. The connectors streamline authentication and eliminate security risks associated with traditional user-generated passwords by automatically generating tokens for each connection, ensuring valid tokens are always used while maintaining full compatibility with existing Npgsql and SQLx features.

The connectors handle IAM token generation, SSL configuration, and connection pooling, enabling customers to scale from simple scripts to production workloads without changing their authentication approach. They also provide opt-in optimistic concurrency control (OCC) retry with exponential backoff, custom IAM credential providers, and AWS profile support, making it easier to develop client retry logic and manage AWS credentials.

To get started, visit the Connectors for Aurora DSQL documentation page. For code examples, visit our GitHub pages for the .NET connector and Rust connector. Get started with Aurora DSQL for free with the AWS Free Tier. To learn more about Aurora DSQL, visit the webpage.

 

​Today we are announcing the release of Aurora DSQL connectors for .NET (Npgsql) and Rust (SQLx) that make it easy to build .NET and Rust applications on Aurora DSQL. The connectors streamline authentication and eliminate security risks associated with traditional user-generated passwords by automatically generating tokens for each connection, ensuring valid tokens are always used while maintaining full compatibility with existing Npgsql and SQLx features. The connectors handle IAM token generation, SSL configuration, and connection pooling, enabling customers to scale from simple scripts to production workloads without changing their authentication approach. They also provide opt-in optimistic concurrency control (OCC) retry with exponential backoff, custom IAM credential providers, and AWS profile support, making it easier to develop client retry logic and manage AWS credentials. To get started, visit the Connectors for Aurora DSQL documentation page. For code examples, visit our GitHub pages for the .NET connector and Rust connector. Get started with Aurora DSQL for free with the AWS Free Tier. To learn more about Aurora DSQL, visit the webpage.  

AWS Transform custom now offers seven new AWS-managed transformations to help you modernize code at scale. These transformations address common modernization scenarios across multiple languages and frameworks.

Generally available transformations includes comprehensive codebase analysis transformation, enabling you to generate hierarchical, cross-referenced documentation covering architecture, business logic, and technical debt, with actionable insights on outdated components and maintenance concerns. The Node.js version upgrade transformation is now generally available and includes comprehensive library upgrade support, enabling you to upgrade Node.js applications from any source version to any target version with full dependency modernization.

Those transformations available in early access include Java performance optimization transformation helps you analyze Java Flight Recorder (JFR) profiling data to detect CPU and memory hotspots and anti- patterns, then applies targeted code fixes to reduce resource usage and improve efficiency. The Log4j to SLF4J migration transformation allows you to remediate Log4j logging dependencies by migrating to the SLF4J logging framework. Also available in early access is the Angular to React migration transformation transforms Angular applications to React. The Angular version upgrade transformation enables you to upgrade Angular applications to the latest version. Finally, the Vue version upgrade transformation upgrades your Vue.js applications to the latest version.

AWS-managed transformations are validated by AWS and can be customized to meet your organization’s specific requirements. All transformations benefit from continual learning, automatically improving quality from every execution.

To get started, install the AWS Transform CLI and run atx custom def list to see all available transformations. For more information, see AWS-Managed Transformations.
AWS Transform custom is available in US East (N. Virginia) and Europe (Frankfurt).

 

​AWS Transform custom now offers seven new AWS-managed transformations to help you modernize code at scale. These transformations address common modernization scenarios across multiple languages and frameworks. Generally available transformations includes comprehensive codebase analysis transformation, enabling you to generate hierarchical, cross-referenced documentation covering architecture, business logic, and technical debt, with actionable insights on outdated components and maintenance concerns. The Node.js version upgrade transformation is now generally available and includes comprehensive library upgrade support, enabling you to upgrade Node.js applications from any source version to any target version with full dependency modernization. Those transformations available in early access include Java performance optimization transformation helps you analyze Java Flight Recorder (JFR) profiling data to detect CPU and memory hotspots and anti- patterns, then applies targeted code fixes to reduce resource usage and improve efficiency. The Log4j to SLF4J migration transformation allows you to remediate Log4j logging dependencies by migrating to the SLF4J logging framework. Also available in early access is the Angular to React migration transformation transforms Angular applications to React. The Angular version upgrade transformation enables you to upgrade Angular applications to the latest version. Finally, the Vue version upgrade transformation upgrades your Vue.js applications to the latest version. AWS-managed transformations are validated by AWS and can be customized to meet your organization’s specific requirements. All transformations benefit from continual learning, automatically improving quality from every execution.
To get started, install the AWS Transform CLI and run atx custom def list to see all available transformations. For more information, see AWS-Managed Transformations. AWS Transform custom is available in US East (N. Virginia) and Europe (Frankfurt).  

AWS Private Certificate Authority (AWS Private CA) now publishes certificate authority (CA) utilization metrics to Amazon CloudWatch, providing visibility into your CA usage. AWS Private CA enforces service quota limits on the number of certificates a CA can issue and the number of CAs you can create per Region. The new metrics track the number of certificates issued by each CA and the total number of CAs in each Region, enabling you to monitor usage against these quotas and proactively manage CA lifecycle to maintain high availability.

With these metrics, you can configure CloudWatch alarms to prevent quota-related service disruptions. For example, you can set alarms to trigger automation that replaces a CA approaching its certificate issuance quota and transitions certificate issuance to a new CA. This is particularly important when using AWS services that rely on AWS Private CA certificates, such as Amazon EKS, Amazon ECS Service Connect, and Amazon WorkSpaces.

The utilization metrics are available in all AWS Regions where AWS Private CA is available. To learn more about AWS Private CA metrics, see the AWS Private CA User Guide. 

 

​AWS Private Certificate Authority (AWS Private CA) now publishes certificate authority (CA) utilization metrics to Amazon CloudWatch, providing visibility into your CA usage. AWS Private CA enforces service quota limits on the number of certificates a CA can issue and the number of CAs you can create per Region. The new metrics track the number of certificates issued by each CA and the total number of CAs in each Region, enabling you to monitor usage against these quotas and proactively manage CA lifecycle to maintain high availability. With these metrics, you can configure CloudWatch alarms to prevent quota-related service disruptions. For example, you can set alarms to trigger automation that replaces a CA approaching its certificate issuance quota and transitions certificate issuance to a new CA. This is particularly important when using AWS services that rely on AWS Private CA certificates, such as Amazon EKS, Amazon ECS Service Connect, and Amazon WorkSpaces.
The utilization metrics are available in all AWS Regions where AWS Private CA is available. To learn more about AWS Private CA metrics, see the AWS Private CA User Guide.   

Amazon CloudWatch Logs Insights now supports a new lookup command that enables customers to enrich log query results with data from reference tables. Developers, DevOps engineers, and SREs working with complex distributed systems often encounter logs containing opaque identifiers such as GUIDs, IP addresses, or internal resource IDs that are difficult to interpret without additional context.

With the lookup command, you can join log data against a lookup table at query time, automatically enriching your results with meaningful values. For example, you can translate a customer ID into a customer name or map an internal IP address to the team that owns it. The new command makes log analysis faster and more intuitive without requiring pre-processing pipelines. 

The lookup command is available today in all commercial AWS Regions. 

To get started, upload a CSV file by navigating to CloudWatch → Settings → Logs.  Next, use the lookup command in your Logs Insights queries by specifying a log field, a lookup table name, and one or more columns.  CSV data does not count toward CloudWatch Logs Insights per GB of data scanned query charges.  To learn more, see the CloudWatch Logs Insights documentation. 

 

​Amazon CloudWatch Logs Insights now supports a new lookup command that enables customers to enrich log query results with data from reference tables. Developers, DevOps engineers, and SREs working with complex distributed systems often encounter logs containing opaque identifiers such as GUIDs, IP addresses, or internal resource IDs that are difficult to interpret without additional context.
With the lookup command, you can join log data against a lookup table at query time, automatically enriching your results with meaningful values. For example, you can translate a customer ID into a customer name or map an internal IP address to the team that owns it. The new command makes log analysis faster and more intuitive without requiring pre-processing pipelines. 
The lookup command is available today in all commercial AWS Regions. 
To get started, upload a CSV file by navigating to CloudWatch → Settings → Logs.  Next, use the lookup command in your Logs Insights queries by specifying a log field, a lookup table name, and one or more columns.  CSV data does not count toward CloudWatch Logs Insights per GB of data scanned query charges.  To learn more, see the CloudWatch Logs Insights documentation.   

Amazon Relational Database Service (Amazon RDS) for Oracle is now available on AWS Outposts. AWS Outposts is a fully managed service that extends AWS infrastructure, AWS services, APIs, and tools to virtually any data center, co-location space, or on-premises facility for a consistent hybrid experience. As a result, customers can run applications using AWS features and services in their on-premises environment for applications that require data residency, regulatory, or other business constraints. With Amazon RDS for Oracle on AWS Outposts, customers can now use a managed Oracle database service on premises, just as they do in the cloud.

Amazon RDS for Oracle on AWS Outposts offers fully managed database management experience such as automated backups, automated patching, point-in-time recovery, monitoring with Amazon CloudWatch, and data encryption at rest with AWS KMS. Amazon RDS for Oracle on AWS Outposts supports multi-AZ deployments across two different Outposts racks for high availability, providing automatic failover to ensure business continuity. For disaster recovery, customer can either restore the database instance in the parent AWS Region using a snapshot taken from the database instance running on AWS Outposts or set up a replica instance in the different Outpost rack or in the parent AWS Region.

Customers can deploy Oracle Database 19c and 21c Enterprise Edition (EE) and Standard Edition 2 (SE2) using the Bring Your Own License (BYOL) model in Amazon RDS for Oracle on AWS Outposts. To get started with RDS for Oracle on Outposts, visit the Amazon RDS on AWS Outposts User Guide. Amazon RDS for Oracle on AWS Outposts is available in all AWS Regions where Amazon RDS on AWS Outposts is available. For pricing information, visit the Amazon RDS on AWS Outposts pricing page.

 

​Amazon Relational Database Service (Amazon RDS) for Oracle is now available on AWS Outposts. AWS Outposts is a fully managed service that extends AWS infrastructure, AWS services, APIs, and tools to virtually any data center, co-location space, or on-premises facility for a consistent hybrid experience. As a result, customers can run applications using AWS features and services in their on-premises environment for applications that require data residency, regulatory, or other business constraints. With Amazon RDS for Oracle on AWS Outposts, customers can now use a managed Oracle database service on premises, just as they do in the cloud. Amazon RDS for Oracle on AWS Outposts offers fully managed database management experience such as automated backups, automated patching, point-in-time recovery, monitoring with Amazon CloudWatch, and data encryption at rest with AWS KMS. Amazon RDS for Oracle on AWS Outposts supports multi-AZ deployments across two different Outposts racks for high availability, providing automatic failover to ensure business continuity. For disaster recovery, customer can either restore the database instance in the parent AWS Region using a snapshot taken from the database instance running on AWS Outposts or set up a replica instance in the different Outpost rack or in the parent AWS Region. Customers can deploy Oracle Database 19c and 21c Enterprise Edition (EE) and Standard Edition 2 (SE2) using the Bring Your Own License (BYOL) model in Amazon RDS for Oracle on AWS Outposts. To get started with RDS for Oracle on Outposts, visit the Amazon RDS on AWS Outposts User Guide. Amazon RDS for Oracle on AWS Outposts is available in all AWS Regions where Amazon RDS on AWS Outposts is available. For pricing information, visit the Amazon RDS on AWS Outposts pricing page.  

We’re announcing availability changes to the following AWS services and features.

Services moving to Maintenance

Services moving to maintenance will no longer be accessible to new customers starting April 30, 2026. Customers already using these services and features can continue to do so. AWS will continue to operate and support these services and features. We recommend that customers learn about the changes in the product pages and documentation.

• Amazon Application Recovery Controller (ARC) – Readiness Check Feature
• Amazon Comprehend – Topic Modeling, Event Detection, and Prompt Safety Classification Features
• Amazon Rekognition – Streaming Events and Batch Image Content Moderation Features
• Amazon Simple Notification Service (SNS) – Message Data Protection (MDP) Feature
• AWS App Runner
• AWS Audit Manager
• AWS CloudTrail Lake
• AWS Glue – Ray Jobs Feature
• AWS IoT FleetWise

Services entering Sunset

The following services are entering sunset, and we are announcing the date upon which we will end operations and support of the service. Customers using these services should click on the links below to understand the sunset timeline and begin planning migration to alternatives as recommended in the updated service web pages and documentation.

• Amazon RDS Custom for Oracle
• Amazon WorkMail
• Amazon WorkSpaces Thin Client
• AWS Service Management Connector

Services reaching End of Support

The following feature has reached end of support and is no longer available as of March 31, 2026.

• Amazon Chime SDK – Proxy Sessions

For customers affected by these changes, we’ve prepared comprehensive migration guides, and our support teams are ready to assist with your transition. Visit AWS Product Lifecycle Page to learn more, and subscribe to the RSS feed for future updates. 

 

​We’re announcing availability changes to the following AWS services and features. Services moving to Maintenance
Services moving to maintenance will no longer be accessible to new customers starting April 30, 2026. Customers already using these services and features can continue to do so. AWS will continue to operate and support these services and features. We recommend that customers learn about the changes in the product pages and documentation.

Amazon Application Recovery Controller (ARC) – Readiness Check Feature
Amazon Comprehend – Topic Modeling, Event Detection, and Prompt Safety Classification Features
Amazon Rekognition – Streaming Events and Batch Image Content Moderation Features
Amazon Simple Notification Service (SNS) – Message Data Protection (MDP) Feature
AWS App Runner
AWS Audit Manager
AWS CloudTrail Lake
AWS Glue – Ray Jobs Feature
AWS IoT FleetWise

Services entering Sunset
The following services are entering sunset, and we are announcing the date upon which we will end operations and support of the service. Customers using these services should click on the links below to understand the sunset timeline and begin planning migration to alternatives as recommended in the updated service web pages and documentation.

Amazon RDS Custom for Oracle
Amazon WorkMail
Amazon WorkSpaces Thin Client
AWS Service Management Connector

Services reaching End of Support
The following feature has reached end of support and is no longer available as of March 31, 2026.

Amazon Chime SDK – Proxy Sessions

For customers affected by these changes, we’ve prepared comprehensive migration guides, and our support teams are ready to assist with your transition. Visit AWS Product Lifecycle Page to learn more, and subscribe to the RSS feed for future updates.   

Amazon ECS Managed Instances now supports Amazon EC2 instance store volumes as a data volume option for container workloads. You can now leverage instance store volumes on your ECS container instances instead of provisioning an Amazon EBS data volume, reducing storage costs and accelerating I/O performance for latency-sensitive workloads.

Amazon ECS Managed Instances is a fully managed compute option designed to eliminate infrastructure management overhead, dynamically scale EC2 instances to match your workload requirements, and continuously optimize task placement to reduce infrastructure costs. With today’s launch, you can enable local storage by configuring a custom ECS Managed Instances capacity provider and selecting the desired Amazon EC2 instance types that include instance store volumes. When an instance lacks instance store volumes or when local storage is disabled, Amazon ECS automatically provisions an Amazon EBS data volume.

Support for instance store is available in all commercial AWS Regions where Amazon ECS Managed Instances is available. To learn more about local storage support, visit the documentation. To learn more about Amazon ECS Managed Instances, visit the feature page, documentation, and AWS News launch blog.

 

​Amazon ECS Managed Instances now supports Amazon EC2 instance store volumes as a data volume option for container workloads. You can now leverage instance store volumes on your ECS container instances instead of provisioning an Amazon EBS data volume, reducing storage costs and accelerating I/O performance for latency-sensitive workloads. Amazon ECS Managed Instances is a fully managed compute option designed to eliminate infrastructure management overhead, dynamically scale EC2 instances to match your workload requirements, and continuously optimize task placement to reduce infrastructure costs. With today’s launch, you can enable local storage by configuring a custom ECS Managed Instances capacity provider and selecting the desired Amazon EC2 instance types that include instance store volumes. When an instance lacks instance store volumes or when local storage is disabled, Amazon ECS automatically provisions an Amazon EBS data volume. Support for instance store is available in all commercial AWS Regions where Amazon ECS Managed Instances is available. To learn more about local storage support, visit the documentation. To learn more about Amazon ECS Managed Instances, visit the feature page, documentation, and AWS News launch blog.  

Now generally available, AWS DevOps Agent is your always-available operations teammate that resolves and proactively prevents incidents, optimizes application reliability and performance, and handles on-demand SRE tasks across AWS, multicloud, and on-prem environments. Building on the preview launch, DevOps Agent now adds new use cases, broader integrations, enhanced intelligence, and enterprise-ready features, including the ability to investigate applications in Azure and on-prem environments, add custom agent skills to extend capabilities, and create custom charts and reports for deeper operational insights.

DevOps Agent investigates incidents and identifies operational improvements as an experienced teammate would: by learning your applications and their relationships, working with your observability tools, runbooks, code repositories, and CI/CD pipelines, and correlating telemetry, code, and deployment data. It autonomously triages incidents and guides teams to rapid resolution, reducing mean time to resolution (MTTR) from hours to minutes, while analyzing patterns across historical incidents to deliver actionable recommendations that prevent future outages.

For the full list of AWS Regions where AWS DevOps Agent is available, visit the Regions list. Pricing details are available on the AWS DevOps Agent pricing page. AWS Support customers receive monthly DevOps Agent credits based on the prior month’s gross AWS Support spend: 100% for Unified Operations, 75% for Enterprise Support, or 30% for Business Support+. For many customers, this significantly reduces or eliminates DevOps Agent costs. For details, visit the support compare page.

If you are a preview customer, review the migration documentation to ensure seamless access to new AWS DevOps Agent capabilities. To learn more, read the launch blog and see getting started.

 

​Now generally available, AWS DevOps Agent is your always-available operations teammate that resolves and proactively prevents incidents, optimizes application reliability and performance, and handles on-demand SRE tasks across AWS, multicloud, and on-prem environments. Building on the preview launch, DevOps Agent now adds new use cases, broader integrations, enhanced intelligence, and enterprise-ready features, including the ability to investigate applications in Azure and on-prem environments, add custom agent skills to extend capabilities, and create custom charts and reports for deeper operational insights. DevOps Agent investigates incidents and identifies operational improvements as an experienced teammate would: by learning your applications and their relationships, working with your observability tools, runbooks, code repositories, and CI/CD pipelines, and correlating telemetry, code, and deployment data. It autonomously triages incidents and guides teams to rapid resolution, reducing mean time to resolution (MTTR) from hours to minutes, while analyzing patterns across historical incidents to deliver actionable recommendations that prevent future outages. For the full list of AWS Regions where AWS DevOps Agent is available, visit the Regions list. Pricing details are available on the AWS DevOps Agent pricing page. AWS Support customers receive monthly DevOps Agent credits based on the prior month’s gross AWS Support spend: 100% for Unified Operations, 75% for Enterprise Support, or 30% for Business Support+. For many customers, this significantly reduces or eliminates DevOps Agent costs. For details, visit the support compare page. If you are a preview customer, review the migration documentation to ensure seamless access to new AWS DevOps Agent capabilities. To learn more, read the launch blog and see getting started.