Skip to content Skip to footer

Release & Changelog Watcher — July 17, 2026

Executive summary

What’s new: AWS announced multiple product updates on 2026-07-16 that improve performance, compliance, cost visibility, observability, and developer experience: Amazon Redshift added Graviton RG sizes rg.large and rg.12xlarge (GA) [9][10]; EC2 High Memory U7in-24TB instances launched in eu-west-3 (Paris) [11]; RDS offers PostgreSQL 19 Beta 2 in the Database Preview Environment [7]; EC2 now surfaces public SSM parameters for public AMIs [5]; S3 Event Notifications include system-generated tags [6]; Amazon Managed Grafana received FedRAMP High in AWS GovCloud (US) [2]; AWS Billing and Cost Management added a Cost Efficiency widget (announced 2026-07-17) to dashboards [3][4]; AWS Sustainability now reports annual water withdrawals by region/service/account (added to the Sustainability console/API) [8]. Separately, OpenAI published an AI ROI scorecard and described teen‑safety features for ChatGPT (policy/metrics updates) [1][12].

Chronological timeline of key developments

  • 2026-07-16: Amazon Redshift general availability — two new Graviton RG instance sizes: rg.large and rg.12xlarge
  • 2026-07-16: Amazon EC2 High Memory U7in-24TB (u7in-24tb.224xlarge) now available in AWS Europe (Paris); 24 TiB DDR5 memory, 896 vCPUs, up to 100 Gbps EBS and 200 Gbps network; optimized for in‑memory DBs (SAP HANA, Oracle, SQL Server) [11].
  • 2026-07-16: Amazon RDS for PostgreSQL 19 Beta 2 released in the RDS Database Preview Environment; features include parallel autovacuum, REPACK CONCURRENTLY, SQL/PGQ (Property Graph Queries), and improved logical replication; preview instances auto-delete after 60 days and snapshots are preview‑bound [7].
  • 2026-07-16: Amazon EC2 now returns the public AWS Systems Manager (SSM) Parameter Store parameter associated with public AMIs in AMI metadata when describing public AMIs — simplifies discovery and use of an alias that resolves to the latest AMI [5].
  • 2026-07-16: Amazon S3 Event Notifications now include system‑generated tags in events delivered to EventBridge, SQS, SNS, and Lambda, enabling cross‑bucket filtering with a single EventBridge rule; available in all Regions at no extra cost [6].
  • 2026-07-16: Amazon Managed Grafana achieved FedRAMP High authorization in AWS GovCloud (US-East and US-West), enabling GovCloud customers and federal/public-sector users to use managed Grafana for metrics and alerting [2].
  • 2026-07-16 / announced 2026-07-17: AWS Billing & Cost Management added a Cost Efficiency widget to BCM Dashboards showing an efficiency score over time with filters by account/region and granularity; integrates with Cost Explorer, Budgets, Savings Plans/RI reports and links to the Cost Optimization Hub; available in all commercial Regions at no additional charge [3][4].
  • 2026-07-16: AWS Sustainability now surfaces annual water withdrawals data by Region, service, and account via the Sustainability console and API (no separate charge) [8].
  • 2026-07-17: OpenAI CFO introduced a practical AI scorecard to measure AI ROI (useful work, cost per successful task, dependability, return on compute) — guidance for measuring AI investments [1].
  • 2026-07-16: OpenAI published materials on making ChatGPT safer for teens — age‑appropriate protections, learning tools, parental controls, expert partnerships (policy/product safety update) [12].

Trends

  • Performance + cost optimization via Graviton: vendors continue to push Arm/Graviton instances for better price/performance (Redshift RG sizes) [9][10].
  • Large-memory specialized instances: High Memory U7 family expands to more regions for in‑memory enterprise databases (U7in-24TB in Paris) [11].
  • Operational discoverability and automation: richer metadata in APIs/events — public AMI SSM parameter in AMI metadata and system tags in S3 events — to simplify automation and scale event rules [5][6].
  • Cost visibility and governance tools: built‑in efficiency metrics (BCM Cost Efficiency widget) and sustainability metrics (water withdrawals) indicate focus on chargeback, optimization, and ESG reporting [3][4][8].
  • Compliance and public sector readiness: FedRAMP High for Managed Grafana shows continued prioritization of GovCloud compliance for observability tools [2].
  • Database innovation adoption path: RDS preview environments enable early testing of major upstream DB versions (PostgreSQL 19 Beta 2) with preview constraints noted [7].
  • Vendor emphasis on AI accountability: measurement frameworks (AI scorecard) and safety features for minors reflect rising focus on ROI metrics and responsible AI [1][12].

Risks

  • Track and compatibility constraints: Redshift RG sizes are limited to current cluster track P202 — customers on P201 must validate upgrade paths and compatibility before migrating [9][10].
  • Preview limitations: PostgreSQL 19 Beta 2 in the preview environment has instance lifetime and snapshot portability limits; not suitable for production validation without full export/import workflows [7].
  • Metadata exposure and governance: exposing public SSM parameters in AMI metadata and including system-generated tags in S3 events may surface information that teams must control to avoid unintended disclosure or misrouting of events [5][6].
  • Migration and performance assumptions: claimed performance and cost improvements (Redshift, Graviton, U7 instances) require workload testing — real gains depend on customer query profiles and application characteristics [9][11].
  • Operational change overhead: adopting new widgets, tags, and parameters requires updates to dashboards, automation, CI/CD, and runbooks; failure to update can cause config drift or missed savings [3][5][6].

Opportunities

  • Right‑size and lower TCO: test Redshift rg.large/rg.12xlarge for cost/perf gains and consider reserved pricing options to reduce costs [9][10].
  • Leverage large‑memory instances: evaluate U7in-24TB for in‑memory DB consolidation or performance boosts for SAP HANA/Oracle workloads in Paris and other regions as U7 expands [11].
  • Streamline AMI management: use public SSM parameter aliases exposed in AMI metadata to automate AMI updates across fleets and reduce manual parameter discovery [5].
  • Simplify event architectures: use system-generated tags in S3 Event Notifications to reduce the number of EventBridge rules and centralize event routing and filtering [6].
  • Improve cost governance: add the BCM Cost Efficiency widget to dashboards, tie it to Cost Optimization Hub recommendations, and include it in scheduled reports to track optimization initiatives [3][4].
  • Early feature evaluation: trial PostgreSQL 19 Beta 2 in the RDS preview to evaluate SQL/PGQ, parallel autovacuum, and REPACK CONCURRENTLY for migration planning [7].
  • Public sector expansion: GovCloud customers can adopt managed Grafana for compliant observability in federal environments [2].
  • Sustainability reporting: include AWS water withdrawals metrics in sustainability dashboards and internal ESG reporting to refine infrastructure decisions [8].
  • Adopt AI measurement: incorporate AI scorecard metrics (useful work, cost per successful task, dependability, return on compute) into procurement and project KPI frameworks to measure AI ROI [1].

Recommended actions

  • Inventory affected services and prioritize tests: run representative workloads on Redshift RG sizes and EC2 U7 instances in a non‑production environment to validate performance and cost claims before migrating production clusters or instances [9][10][11].
  • Enable and configure new cost tooling: add the BCM Cost Efficiency widget to central cost dashboards, schedule reports, and link to the Cost Optimization Hub to operationalize recommendations [3][4].
  • Update automation and AMI workflows: modify CI/CD and infra-as-code to use the SSM parameter alias surfaced in public AMI metadata for easier AMI updates; validate IAM controls for parameter access [5].
  • Rework event rules to use system tags: consolidate EventBridge rules by switching to system-generated tag filters for S3 notifications where appropriate; test rule behavior to confirm no breakage [6].
  • Plan PostgreSQL 19 evaluation: use the RDS preview for compatibility testing, follow preview constraints (60‑day instance retention, snapshot portability limits), and use pg_dump/pg_restore for data migration tests [7].
  • Confirm compliance posture: GovCloud and public-sector teams should validate Managed Grafana FedRAMP High documentation and engage AWS account teams for deployment guidance [2].
  • Integrate sustainability metrics: add AWS water withdrawals to internal sustainability dashboards and factor water intensity into region/service placement decisions where relevant [8].
  • Adopt AI ROI measurement: pilot the AI scorecard for active AI projects to quantify useful work, compute returns, and dependability; use results to guide investment and product safety choices (including teen‑safety controls where applicable) [1][12].
  • Mitigate risks: document migration paths for Redshift track upgrades, secure metadata exposure (IAM, logging), and update runbooks and team training to reflect new features and defaults [9][5][6].

Sources

  1. [1] A scorecard for the AI age
  2. [2] Amazon Managed Grafana achieves FedRAMP High authorization in AWS GovCloud (US)
  3. [3] Track cost efficiency trends directly in Billing and Cost Management Dashboards with the new Cost Efficiency widget
  4. [4] Track cost efficiency trends directly in Billing and Cost Management Dashboards with the new Cost Efficiency widget
  5. [5] Amazon EC2 now surfaces the public SSM parameters associated with public AMIs
  6. [6] Amazon S3 Event Notifications now include system-generated tags
  7. [7] PostgreSQL 19 Beta 2 is now available in Amazon RDS Database Preview Environment
  8. [8] AWS Sustainability service now includes water withdrawals data
  9. [9] Amazon Redshift adds rg.large and rg.12xlarge instance sizes
  10. [10] Amazon Redshift adds rg.large and rg.12xlarge instance sizes
  11. [11] Amazon EC2 High Memory U7in-24TB instances now available in AWS Europe (Paris) region
  12. [12] Why teens deserve access to safe AI

Leave a comment

0.0/5