As a Cloud Security Architect, DevSecOps Lead, and Terraform Specialist, I've reviewed today's AWS announcements, focusing on their security implications, architectural considerations, and Terraform integration. This summary provides a professional overview of how these updates empower secure, efficient, and compliant cloud operations.
AI and Machine Learning
- AWS Agent Registry - Generally Available: AWS Agent Registry is now generally available, offering a private, governed catalog for AI agents, tools, skills, MCP servers, and custom resources within your organization. This service significantly enhances visibility into an organization's AI landscape, promoting reuse and reducing redundant development. It integrates seamlessly with Amazon Bedrock AgentCore, Amazon Quick, and Kiro IDE. From a security perspective, centralized governance is paramount for managing AI agents, preventing shadow IT, and ensuring compliance. The new capabilities include provisioning and management as code via AWS CloudFormation, Terraform, and AWS CDK, comprehensive tagging for organization, cost, and access control, and secure sharing via AWS Resource Access Manager (RAM) for cross-account and organization-wide registries. *Terraform Support*: Supported via
aws_agent_registryresource in Terraform AWS Provider: aws_agent_registry in v5.71.0+. - Amazon Quick - AWS Agent Registry Integration: Amazon Quick now directly integrates with the AWS Agent Registry, allowing users to search, browse, and enable available agents and MCP servers from their organization's registry within the Quick admin console. This streamlines the deployment and accessibility of AI agents for business users, leveraging existing Bedrock AgentCore implementations. For security, this integration promotes the use of approved, governed agents, enhancing control and reducing the risk of unauthorized AI tool usage within business intelligence workflows. *Terraform Support*: Pending provider release (use AWS CLI/SDK or custom provisioner in the interim). This feature is an integration, not a direct configurable resource in QuickSight itself, and the underlying Agent Registry resource is separate.
Analytics
- Amazon Redshift - Apache Iceberg v3 Support: Amazon Redshift now supports reading from and writing to Apache Iceberg v3 tables in your data lake. This update introduces support for default column values, row lineage, and deletion vectors. These features simplify schema evolution, enable the creation of robust incremental pipelines and Change Data Capture (CDC) workflows, and deliver faster reads and writes for high-frequency update and delete workloads. From a security and compliance standpoint, row lineage is invaluable for auditing and data governance, providing clear traceability of data changes. Deletion vectors enhance the efficiency of data removal, critical for compliance with data privacy regulations like GDPR and CCPA. *Terraform Support*: The
aws_redshift_clusterresource manages Redshift clusters, and Iceberg v3 support is a feature of the Redshift engine. Table properties likeformat-versionare typically managed via SQL commands. While the cluster itself is managed by Terraform, direct Terraform arguments for Iceberg v3 format version are not explicitly exposed. However, the underlying Redshift cluster resource is supported in Terraform AWS Provider: aws_redshift_cluster in v5.40.0+. - Amazon Redshift - IAM Identity Center Authentication with Enhanced VPC Routing (EVR): Amazon Redshift now supports AWS IAM Identity Center authentication for provisioned clusters and serverless workgroups configured with Enhanced VPC Routing (EVR). This enables single sign-on (SSO) with corporate credentials, ensuring that all authentication traffic traverses your Amazon VPC and remains on the AWS network via AWS PrivateLink interface VPC endpoints. This is a significant security enhancement, as it enforces SSO, isolates authentication traffic within the VPC, and allows for granular governance using security groups, network ACLs, and endpoint policies. It also supports IAM Identity Center multi-Region replication for distributed Redshift deployments. *Terraform Support*: Supported via arguments like
iam_identity_center_configurationwithin Terraform AWS Provider: aws_redshift_cluster oraws_redshiftserverless_workgroupin v5.68.0+. - Amazon OpenSearch Service - Expanded Cluster Insights: Amazon OpenSearch Service's Cluster Insights now includes 17 new diagnostic insights, with 6 critical-severity insights designed to detect conditions leading to a Red cluster status (primary shard unassignment) and 11 insights (Medium to Critical) addressing issues preventing replica shard allocation (Yellow cluster status). These insights cover a broad spectrum of problems, including JVM out-of-memory errors, sustained CPU saturation, and zone imbalance, providing tailored remediation recommendations. This operational enhancement is crucial for maintaining the availability and performance of OpenSearch clusters, which are often foundational for security monitoring, logging, and incident response systems. *Terraform Support*: Pending provider release (this is an operational enhancement, not a direct configurable resource via Terraform at this time). The underlying
aws_opensearch_domainresource is supported, but these specific insights are service-managed.
Compute
- Amazon EC2 - R9g and R9gd Memory-Optimized Instances: AWS has announced the release of Amazon EC2 R9g and R9gd instances, powered by AWS Graviton5 processors. These instances deliver best-in-class price-performance for memory-intensive workloads, offering up to 25% better compute performance and significantly faster speeds for databases, web applications, and machine learning compared to Graviton4-based instances. R9gd instances further include local NVMe-based SSD block-level storage. A key security highlight is their foundation on the sixth-generation AWS Nitro System, featuring the Nitro Isolation Engine, which provides mathematically verified isolation for customer workloads, significantly enhancing data protection and integrity for highly sensitive applications. *Terraform Support*: Supported via the
instance_typeargument in Terraform AWS Provider: aws_instance in v5.20.0+. - AWS Lambda - Recursive Loop Detection Expansion: AWS Lambda has expanded its recursive loop detection feature to all commercial AWS Regions. This automatic guardrail is designed to detect and halt recursive invocation patterns that can arise from misconfigurations or code defects (e.g., a Lambda function inadvertently triggering itself via an event source). This mechanism prevents unintended resource usage and potential billing spikes. Upon detection, Lambda automatically stops processing the event and notifies users via the AWS Health Dashboard. This feature is a critical security control against denial-of-service (DoS) scenarios caused by misconfigured event sources, helping to maintain the stability and cost-efficiency of serverless applications. It is enabled by default for functions using a supported SDK version and can be optionally disabled via the
PutFunctionRecursionConfigAPI for intentional recursive patterns. *Terraform Support*: Pending provider release (thePutFunctionRecursionConfigAPI is new, and Terraform support for this specific configuration is not yet available foraws_lambda_function).
Database
- Amazon Timestream for InfluxDB - Regional Expansion: Amazon Timestream for InfluxDB is now available in 8 new AWS Regions. This expansion enables customers in these geographies to deploy and manage InfluxDB instances with AWS's operational benefits, including Multi-AZ high availability for resilience, support for read replicas to enhance read throughput and disaster recovery, enhanced durability for data integrity, and flexible multi-node scaling. From a security and compliance standpoint, this regional availability significantly improves data residency options, allowing organizations to meet local regulatory requirements, while Multi-AZ and read replicas bolster resilience and data availability, crucial for maintaining a strong security posture. *Terraform Support*: Supported in Terraform AWS Provider: aws_timestream_influxdb_instance in v5.30.0+.
- Amazon Aurora Serverless - Performance and Regional Expansion: Amazon Aurora Serverless now delivers up to 30% better performance and enhanced scaling capabilities in 5 additional AWS Regions (Asia Pacific (New Zealand), Asia Pacific (Thailand), Africa (Cape Town), Europe (Milan), and Mexico (Central)) for both Aurora PostgreSQL and Aurora MySQL. These improvements, delivered in platform version 4 at no additional cost, efficiently handle workloads where multiple tasks compete for resources, such as busy web applications and agentic AI applications with traffic bursts and idle periods. The regional expansion supports data residency requirements, while enhanced performance and scaling contribute to the overall resilience and availability of database services, indirectly bolstering security by ensuring systems can handle legitimate load and resist certain types of availability attacks. *Terraform Support*: Supported via
aws_rds_clusterandaws_rds_cluster_instanceresources for Aurora Serverless v2, with regional availability implicit. The performance enhancements are platform-level. Terraform AWS Provider: aws_rds_cluster in v4.20.0+.
End User Computing
- Amazon WorkSpaces Applications - Regional Expansion: Amazon WorkSpaces Applications has expanded its availability to three new AWS Regions: Europe (Zurich), Asia Pacific (Osaka), and Canada West (Calgary). This expansion allows customers in these geographical areas to leverage the service for application and desktop streaming, benefiting from reduced latency for end-users and enhanced capabilities for meeting local data residency and compliance mandates. Localized deployments are a key architectural and security consideration, as they can minimize data transit risks and help adhere to specific regional regulatory frameworks. *Terraform Support*: Supported via
aws_appstream_stackand related resources. Regional availability is implicit with provider version support for the resource itself. Terraform AWS Provider: aws_appstream_stack in v2.0.0+.
Networking & Content Delivery
- AWS Interconnect - Multicloud (Microsoft Azure Preview): AWS Interconnect - multicloud is now in public preview for Microsoft Azure, complementing existing general availability for OCI and Google Cloud. This service enables direct, private, and scalable network connections between AWS and other cloud service providers, based on an open specification for network interoperability. From a security perspective, establishing dedicated, private network paths significantly reduces exposure to the public internet, enhancing data confidentiality and integrity for multicloud architectures. This is critical for organizations with stringent compliance requirements for cross-cloud data transfer and for building robust, high-performance multicloud networking strategies. *Terraform Support*: Pending provider release (as a new service in preview, direct Terraform support is not yet available).
Security, Identity, and Compliance
- Amazon Cognito - GetClientToken API for Machine-to-Machine (M2M) Authorization: Amazon Cognito now supports the
GetClientTokenAPI operation, enabling app clients to obtain access tokens for machine-to-machine (M2M) authorization directly through the AWS SDK, CLI, or API, without configuring a user pool domain. This new API allows app clients to authenticate with their client ID and secret to receive access tokens authorized for custom scopes on your resource servers. A significant security advantage is its seamless integration with AWS WAF and VPC interface endpoints (AWS PrivateLink), allowing for stricter network controls and reducing reliance on public endpoints for M2M communication. This provides a more native and secure method for AWS-hosted applications to perform M2M authentication. *Terraform Support*: While theGetClientTokenAPI is new, the underlyingaws_cognito_user_pool_clientresource and its configuration (e.g., client ID, secret, resource servers, scopes) are managed by Terraform AWS Provider: aws_cognito_user_pool_client in v2.0.0+. The API itself is an operational method. - Partner Revenue Measurement - User Agent String Expansion: The Partner Revenue Measurement User Agent string now supports additional AWS services. Partners who have already embedded the user agent (format
APN_1.1/pc_<AWS Marketplace product-code>$) in their applications automatically benefit from expanded coverage with no additional implementation needed. The additional service consumption measured through this expansion is now visible in the Attributed Revenue Dashboard in AWS Partner Central. While not a direct security feature, improved visibility into service consumption can aid in cost allocation, identify potential unauthorized usage patterns when integrated with security analytics, and enhance compliance for financial reporting and partner program adherence. *Terraform Support*: While the user agent string itself is embedded in applications, the underlying AWS services and resources that generate this revenue can be provisioned and managed using Terraform AWS Provider in v4.67.0+.
Announcements in this summary
- Amazon Quick Analytics
- Amazon Timestream for InfluxDB Database
- Amazon Redshift Data Warehouse
- Amazon EC2 Compute
- Partner Revenue Measurement Analytics
- Amazon Redshift Analytics
- Amazon OpenSearch Service Analytics
- AWS Lambda Compute
- Amazon Cognito Security, Identity, and Compliance
- AWS Interconnect - multicloud Networking & Content Delivery
- Amazon Aurora Serverless Database
- Amazon WorkSpaces Applications End User Computing
- AWS Agent Registry AI and Machine Learning