8/8/2025, 12:00:00 AM ~ 8/11/2025, 12:00:00 AM (UTC)
Recent Announcements
Amazon CloudWatch RUM is now generally available in 2 additional AWS regions
Amazon CloudWatch RUM, which enables customers to monitor their web applications by collecting client side performance and error data in real time, is additionally available in the following AWS Regions starting today: Asia Pacific (Thailand), and Mexico (Central).\n CloudWatch RUM provides curated dashboards for web application performance experienced by real end users including anomalies in page load steps, core web vitals, and JavaScript and HTTP errors across different geolocations, browsers, and devices. Custom events and metrics sent to CloudWatch RUM can be easily configured to monitor specific parts of the application for real user interactions, troubleshoot issues, and get alerted for anomalies. CloudWatch RUM comes integrated with the application performance monitoring (APM) capability, CloudWatch Application Signals. As a result, client-side data from your application can easily be correlated with performance metrics such as errors, faults, and latency observed in your APIs (service operations) and dependencies to address the root cause. To get started, see the RUM User Guide. Usage of CloudWatch RUM is charged on the number of collected RUM events, which refers to each data item collected by the RUM web client, as detailed here.
OpenSearch UI supports Fine Grained Access Control by SAML attributes
Amazon OpenSearch Service now supports Fine Grained Access Control (FGAC) for OpenSearch UI when accessed through SAML via IAM federated. OpenSearch UI is the unified interface for observability and security analytics on Amazon OpenSearch Service. SAML via IAM federated is a popular choice to enable Identity Provider (IdP) initiated Single Sign-On experience for accessing OpenSearch UI. FGAC enables you to define precise data access control based on user attributes provided from your IdP during SAML authentication and authorization. This level of dynamic and granular access control is crucial for multi-tenant deployments and meeting data governance requirement in regulated industries. \n With FGAC support, you can now configure attribute mappings from IdP user roles and attributes to OpenSearch backend roles. These roles can be scoped to specific OpenSearch domains and serverless collections, allowing you to define index-level permissions and document-level security for more granular data access controls. You can easily manage users and groups within your existing IdP, and OpenSearch data source permissions are automatically applied based on the user’s SAML assertion, reducing administrative friction. Furthermore, audit trails become clearer as user actions are tied not just to IAM roles but to SAML attributes, simplifying data governance. FGAC is an optional feature for SAML via IAM federated. It is available in all regions that OpenSearch UI is available. To learn more, check: OpenSearch UI dev guide.
Amazon SageMaker HyperPod now supports continuous provisioning for enhanced cluster operations
Amazon SageMaker HyperPod now offers continuous provisioning, a new capability that enables greater flexibility and efficiency for enterprise customers running large-scale AI/ML workloads. AI/ML customers need to start training quickly, scale seamlessly, perform maintenance without disrupting operations, and have granular visibility into cluster operations. Customers also require the ability to efficiently manage dynamic inference workloads where capacity needs change frequently, making operational agility critical for successful AI initiatives.\n With continuous provisioning, SageMaker HyperPod automatically provisions remaining capacity in the background while training jobs can begin immediately on available instances. HyperPod will retry in the background when it encounters node provisioning failures and ensure clusters reliably reach their desired scale without requiring any manual intervention. This helps customers reduce time-to-training and maximizes resource utilization across dynamic workloads. You can now perform concurrent operations such as scaling nodes independently, applying patches, or adjusting different instance groups simultaneously, thus increasing efficiency. The enhanced event-driven architecture provides comprehensive real-time visibility through the new Events APIs, offering complete operational history to enable faster troubleshooting and better decision-making. These capabilities enable customers to achieve improved operational agility, better resource utilization, and enhanced visibility into cluster operations, allowing AI/ML teams to focus on innovation rather than infrastructure management. This feature is currently available for SageMaker HyperPod clusters using the EKS orchestrator. You can enable continuous provisioning by setting the NodeProvisioningMode parameter to “Continuous” when creating new HyperPod clusters using the CreateCluster API. This feature is available in all AWS Regions where Amazon SageMaker HyperPod is supported. To learn more about continuous provisioning, see the Amazon SageMaker HyperPod User Guide.
Amazon ECS console now supports real-time log analytics via Amazon CloudWatch Logs Live Tail
The Amazon ECS console now natively integrates with Amazon CloudWatch Logs Live Tail, enabling real-time log streaming directly within the ECS console.\n ECS customers often need to analyze logs in real time to troubleshoot application issues, investigate deployment failures, and monitor container health. Previously, customers had to navigate away from the ECS console to the CloudWatch console to access real-time log streams, creating workflow interruptions. With this integration, ECS customers can now monitor and troubleshoot containerized applications in real-time without switching between AWS consoles. To start tailing logs in the ECS console, navigate to the logs tab on any ECS service or task details page. Simply click the “Open CloudWatch Logs Live Tail” button, and click “Start”. The Live Tail panel remains visible as you navigate the console, enabling log monitoring while checking other operational metrics. Live Tail in the ECS console is now available in all AWS commercial regions. To learn more, visit the ECS developer guide.
The Amazon SageMaker lakehouse architecture now automates optimization of Apache Iceberg tables stored in Amazon S3 with catalog-level configuration, reducing metadata overhead and improving query performance. Previously, optimizing Iceberg tables in AWS Glue Data Catalog required updating configurations for each table individually. Now, you can enable automatic optimization for new Iceberg tables with a one-time Data Catalog configuration. Once enabled, for any new table or updated table, Data Catalog continuously optimizes tables by compacting small files, removing snapshots, and unreferenced files that are no longer needed, resulting in controlled storage costs and faster queries.\n You can get started by selecting the default catalog in the AWS Lake Formation console and enabling optimizations in the table optimizations configuration tab. You have the choice of additional granular control at the table configuration level, such as sort/z-order compaction strategy, thresholds for the number of small files to trigger compaction, intervals between consecutive snapshot expirations, and unreferenced data cleanup operations.
This feature is available through the AWS Management Console, AWS CLI, and AWS SDKs in 15 AWS Regions: US East (N. Virginia, Ohio), US West (Oregon), Canada (Central), Europe (Ireland, London, Frankfurt, Stockholm), Asia Pacific (Tokyo, Seoul, Mumbai, Singapore, Sydney, Jakarta), and South America (São Paulo). To learn more, read the blog, and visit the Data Catalog documentation.
AWS Blogs
AWS Japan Blog (Japanese)
- The AI-Driven Development Lifecycle: Reinventing Software Engineering
- Automating the startup and shutdown of SAP NetWeaver ABAP on SAP HANA with AWS Systems Manager for SAP
- PostgreSQL performance improvements: diagnosing and addressing lock manager conflicts
- Understanding Kiro’s pricing: Spec, Vibe, and usage tracking
- Minimize AI halation and achieve verification accuracy of up to 99% using automated inference checks: available now
- OpenAI’s open weight model is now available on AWS
- Introducing Amazon Elastic VMware Service to run VMware Cloud Foundation on AWS
- Introducing Amazon S3 Vectors: First cloud storage with large-scale native vector support (preview)
- Resolve incomplete data using AWS Entity Resolution’s advanced rule-based fuzzy matching