As a Cloud Security Architect, DevSecOps Lead, and Terraform Specialist, here's your comprehensive daily summary of the latest AWS features and service updates, with critical insights into their security implications and Terraform availability.
Compute
- Amazon EC2: ImdsV1Enabled CloudWatch Metric
Amazon EC2 now provides an ImdsV1Enabled CloudWatch metric, allowing you to track which EC2 instances have Instance Metadata Service Version 1 (IMDSv1) enabled. This feature is crucial for identifying instances that need to be migrated to the more secure IMDSv2, eliminating the need for custom tooling. This enhances your security posture by making it easier to enforce IMDSv2, which mitigates Server-Side Request Forgery (SSRF) and other vulnerabilities related to metadata access. Architecturally, this metric provides immediate visibility into compliance gaps, enabling proactive remediation and strengthening your overall cloud security framework. *Terraform Support*: While the metric itself is a reporting feature, the underlying EC2 instance metadata service configuration is managed by aws_instance with the metadata_options block (specifically http_tokens = "required" for IMDSv2 enforcement) in Terraform AWS Provider: aws_instance in v4.0.0+.
- AWS Lambda: Durable Functions Default Quotas Increase
AWS Lambda has significantly increased default quotas for durable functions, boosting limits for running durable executions, function invocation rates, and various API rates (e.g., CheckpointDurableExecution, GetDurableExecutionState). These increases enable the development and deployment of larger-scale, more complex, and resilient serverless workflows without immediate concerns about hitting service limits. Architecturally, this empowers developers to build more sophisticated orchestration patterns and long-running processes using Lambda, enhancing the scalability and robustness of serverless applications. While not a direct security feature, increased resilience contributes to operational security by reducing failure points. *Terraform Support*: Default quota increases are service-level enhancements and not directly configurable via Terraform. While aws_service_quota can manage requests for specific quota increases, this particular update is a default service improvement. (Use AWS CLI/SDK or custom provisioner for specific quota requests if needed).
Machine Learning
- Amazon Bedrock: Kimi K3.1 Pro (Highspeed) from Moonshot AI
Moonshot AI's Kimi K3.1 Pro (Highspeed) model is now available on Amazon Bedrock. This open-weight, natively multimodal agentic model boasts an impressive 1-million-token context window and delivers high output tokens per second, making it ideal for fast agentic use cases like coding and document processing. It also supports explicit prompt caching to reduce latency and input costs. Crucially, it operates within the same robust security boundary as proprietary models on Bedrock, ensuring data privacy and compliance with existing Bedrock security controls. Architecturally, this model unlocks new possibilities for advanced AI applications requiring extensive context and rapid processing. *Terraform Support*: The availability of a new model is a service offering, not a direct Terraform resource. However, the underlying Amazon Bedrock service and its security configurations, such as logging and KMS encryption (e.g., aws_bedrock_model_invocation_logging_configuration), are managed by Terraform AWS Provider: aws_bedrock_model_invocation_logging_configuration in v5.26.0+.
Storage
- Amazon FSx for Lustre: 16x Higher Metadata IOPS on Intelligent-Tiering File Systems
Amazon FSx for Lustre now delivers up to 192,000 metadata IOPS (a 16x increase) on Intelligent-Tiering file systems. This significant enhancement accelerates metadata operations such as file/directory creation, listing, reading, and deletion, particularly beneficial for workloads that mix hot and cold data. This directly translates to improved time-to-results for demanding applications like machine learning research and high-performance computing (HPC), without requiring consistent SSD-level performance across all data. While primarily a performance feature, faster processing can indirectly reduce the window of exposure for sensitive data during active operations. *Terraform Support*: The performance improvement is a service-level enhancement. The Amazon FSx for Lustre file system itself, including its deployment_type and per_unit_storage_throughput configurations, is managed by aws_fsx_lustre_file_system in Terraform AWS Provider: aws_fsx_lustre_file_system in v5.82.0+.
- Amazon S3 Tables: Managed Migration from Apache Hive to Apache Iceberg
Amazon S3 Tables now offers managed migration capabilities to convert your Apache Hive tables to Apache Iceberg. This transition brings critical data lake features like ACID transactions, schema evolution, and time travel, significantly enhancing data integrity, reliability, and governance. The service validates, converts, and copies your data while production workloads remain operational, allowing for step-by-step review, synchronization of changes, and rollback options. This is a game-changer for data lake modernization, improving auditability and compliance. Data is recompressed to Zstandard (ZSTD) and compacted for better performance and lower storage costs. *Terraform Support*: This managed migration *process* is not directly exposed as a Terraform resource. While the underlying S3 buckets (aws_s3_bucket) and AWS Glue catalog resources (aws_glue_catalog_table) are fully manageable with Terraform, the migration workflow itself is currently handled via the AWS Console, APIs, or Agent Toolkit. (Use AWS CLI/SDK or custom provisioner in the interim).
Management & Governance
- Amazon CloudWatch: OTLP/HTTP Clients Authenticate with Bearer Token for X-Ray
CloudWatch now supports authentication with a static bearer token via the Authorization header for OTLP/HTTP clients (including OpenTelemetry Collector and SDK exporters) sending traces to the AWS X-Ray OTLP traces endpoint. This is particularly valuable for environments outside of AWS where implementing SigV4 signing can be complex. While SigV4 remains the recommended method for AWS-native workloads, this new option provides greater flexibility for hybrid and multi-cloud observability. Security insight: The bearer token should be managed as a secret, ideally in AWS Secrets Manager, to ensure secure handling and rotation. Architecturally, this simplifies the integration of external OpenTelemetry data sources into AWS's comprehensive observability suite. *Terraform Support*: While aws_secretsmanager_secret can manage the bearer token itself, the specific configuration of the AWS X-Ray OTLP endpoint to accept this bearer token authentication method is not a direct Terraform resource. (Use AWS CLI/SDK or custom provisioner in the interim).
Developer Tools
- AWS CodeConnections: Now Available in AWS GovCloud (US-West)
AWS CodeConnections has expanded its availability to the AWS GovCloud (US-West) Region, complementing its existing presence in GovCloud (US-East). This enables customers operating in GovCloud (US-West) to establish secure connections to their source repositories within the same region as their AWS CodePipeline pipelines. This is a critical update for government and regulated customers, ensuring data residency, compliance with stringent regulatory requirements, and enabling a fully compliant and localized CI/CD workflow within the isolated GovCloud environment. Architecturally, it completes the secure CI/CD toolchain for GovCloud (US-West) deployments. *Terraform Support*: The aws_codestarconnections_connection resource fully supports creating and managing connections, including in GovCloud regions, in Terraform AWS Provider: aws_codestarconnections_connection in v4.0.0+.
Analytics
- Amazon OpenSearch Service: Optimized Engine for OpenTelemetry Logs and Traces
Amazon OpenSearch Service now features an optimized engine that intelligently maps OpenTelemetry logs and traces to object, nested, and flat object fields, preserving the nested structure of span events, span links, prompts, and tool calls. This allows for field-level queries without data flattening during ingestion. Furthermore, on hot-tier data, the engine supports update, delete, and upsert operations by document ID, enabling in-place correction of late or erroneous data. This significantly enhances data integrity and flexibility for observability data. Security insight: The ability to correct data in place improves data quality and auditability, crucial for maintaining accurate security logs. Architecturally, it streamlines observability data ingestion and analysis, providing richer insights for application performance monitoring, service maps, and incident response. These features are available on OpenSearch 3.7. *Terraform Support*: The Amazon OpenSearch Service domain, which hosts this optimized engine, is managed by aws_opensearch_domain in Terraform AWS Provider: aws_opensearch_domain in v5.0.0+. The engine optimization is a service-level feature of OpenSearch 3.7, not a direct configurable argument.
Application Integration
- Amazon Managed Workflows for Apache Airflow (Amazon MWAA): Cost Allocation Tags
Amazon MWAA now supports cost allocation tags, which are automatically propagated to all associated compute resources, including Apache Airflow workers, schedulers, and web servers. This enables organizations to track and analyze their MWAA spending by specific tags (e.g., project, team, cost center) using AWS Cost Explorer and AWS Cost and Usage Reports. This feature is vital for financial governance, chargeback/showback models, and ensuring compliance with internal budgeting and reporting standards. Tags can be configured during environment creation or updates, and their keys must be enabled in the AWS Billing and Cost Management console for activation. *Terraform Support*: Cost allocation tags for MWAA environments are fully supported via the tags argument of the aws_mwaa_environment resource in Terraform AWS Provider: aws_mwaa_environment in v3.20.0+.
Announcements in this summary
- Amazon EC2 Compute
- Amazon Bedrock Machine Learning
- Amazon FSx for Lustre Storage
- AWS Lambda Compute
- Amazon S3 Tables Storage
- Amazon CloudWatch Management & Governance
- AWS CodeConnections Developer Tools
- Amazon OpenSearch Service Analytics
- Amazon Managed Workflows for Apache Airflow (Amazon MWAA) Application Integration