Introduction
AWS keeps pushing the boundaries of cloud innovation, delivering exciting new features and enhancements. Dive into the most impactful updates from April 2026 that every developer and cloud architect needs to know.
Key Updates
1. Amazon EC2 for SQL Server HA now supports health notifications
Published: April 28, 2026
AWS has announced a handy new feature: you can now get notifications via the AWS Health Dashboard if Amazon EC2 for SQL Server fails to detect a valid SQL Server High Availability (HA) status.
This is a game-changer for customers using EC2 SQL HA clusters—registered via AWS Console or CloudFormation—to cut licensing costs. Notifications arrive through AWS Health, Amazon EventBridge events, or email, empowering you to act fast and dodge surprise bills.
For full details, check the High Availability for SQL Server on Amazon EC2 user guide. The feature works in all AWS Regions supporting Amazon EC2 SQL HA, plus AWS GovCloud (US) Regions.
2. Amazon Corretto April 2026 Quarterly Updates
Published: April 22, 2026
On April 22, 2026, Amazon rolled out quarterly security and critical updates for Amazon Corretto’s Long-Term Supported (LTS) and Feature Release (FR) OpenJDK versions. Download Corretto 26.0.1, 25.0.3, 21.0.11, 17.0.19, 11.0.31, and 8u492 now. As a no-cost, multi-platform, production-ready OpenJDK distribution, Corretto is built for reliability.
Note: This marks the final Corretto 8 release with JavaFX binaries—they’ll drop in the July 2026 update. See migration tips on the Corretto 8 GitHub.
Grab the latest from the Corretto home page, or update Linux systems via Apt, Yum, or Apk repos. Feedback always welcome!
3. Amazon SageMaker Unified Studio now supports VPC for notebook kernels
Published: April 22, 2026
Amazon SageMaker Unified Studio now brings Amazon Virtual Private Cloud (Amazon VPC) support to notebook kernels. Kernels now run inside the VPC set at the domain level, delivering network isolation for interactive data and machine learning (ML) workloads—perfect for meeting strict security and compliance needs.
Data engineers, analysts, and scientists can now tap into private resources straight from notebooks. Kernels inherit VPC settings, subnets, and security groups from the domain, letting admins centralize network policies. Query private databases, hit internal APIs, or handle non-public data—all in one SQL, Python, or natural language-powered environment via the built-in data agent.
Important: This applies only to interactive notebook compute (like Python code and dataframes). Check docs for other engines.
Available in all AWS Regions with SageMaker Unified Studio. Learn more in the SageMaker Unified Studio user guide and on the Amazon SageMaker product page.
4. AWS KMS now tracks last usage of all KMS keys
Published: April 16, 2026
AWS Key Management Service (KMS) now shows the last cryptographic operation for your keys—no more manual log digging. Security admins and compliance teams get quick insights into usage via timestamps, operation types, and AWS CloudTrail event IDs, right in the KMS console or APIs.
Spot unused keys for cleanup, confirm active usage, or audit via CloudTrail. Plus, the new kms:TrailingDaysWithoutKeyUsage condition key prevents accidental deletes of recently used keys.
Live in all AWS Regions with KMS, including commercial, GovCloud (US), and China Regions. Details in Determine past usage of a KMS key from the AWS KMS Developer Guide.
5. Amazon CloudWatch adds visual agent configuration to the EC2 console
Published: April 9, 2026
Amazon CloudWatch now offers a visual editor for agent configs directly in the Amazon EC2 console—no JSON tweaking required. The agent gathers metrics, logs, and traces from EC2 instances, feeding them to CloudWatch and AWS X-Ray. Build configs graphically: pick metrics, logs, and targets, then deploy with one click.
Select instances in the EC2 console to install the agent or set tag-based policies for fleet-wide automation. Instance details show agent status, configs, and health troubleshooting. Policies auto-apply to new instances, even in auto-scaling groups.
Start in the EC2 console: pick an instance, hit the EC2 monitoring tab. Available in all AWS Commercial Regions at no extra cost (standard CloudWatch pricing applies).
6. AWS Glue 5.1 is now available in all AWS Commercial and AWS GovCloud (US) Regions
Published: April 9, 2026
AWS Glue 5.1 has expanded to Asia Pacific (New Zealand), AWS GovCloud (US-West), and AWS GovCloud (US-East)—now live across all commercial and GovCloud (US) Regions.
This serverless data integration service makes discovering, preparing, moving, and integrating data a breeze. Version 5.1 upgrades to Apache Spark 3.5.6, Python 3.11, and Scala 2.12.18 for better performance and security. It boosts open table formats: Apache Hudi 1.0.2, Apache Iceberg 1.10.0 (now with v3.0 support for default columns, deletion vectors, multi-arg transforms, and row lineage), and Delta Lake 3.3.2.
New: AWS Lake Formation fine-grained access controls extend to write ops (DML/DDL) for Spark DataFrames/SQL—previously read-only. Full-table access for Hudi/Delta tables adds robust security.
Launch via AWS APIs, CLI, SDK, or Glue Studio. Explore the AWS Glue product page and docs.
7. Amazon SageMaker supports notebooks and data agent for IdC domains
Published: April 8, 2026
Amazon SageMaker Unified Studio now delivers serverless notebooks with a built-in data agent for AWS IAM Identity Center (IdC) domains—previously limited to IAM domains. IdC users get seamless access to high-performance, serverless notebooks for analytics and ML.
One unified spot for SQL queries, Python execution, big data jobs, ML, and visualizations. The AI data agent speeds things up by generating code/SQL from natural language and guiding workflows. Mix SQL, Python, and prompts freely—no tool-switching hassles. Example: Explore with SQL, analyze with Python/ML, or auto-generate via prompts. Powered by Amazon Athena for Spark, it scales from queries to petabyte processing.
Available wherever SageMaker Unified Studio runs. See the SageMaker notebooks user guide and data agent user guide.
8. Paraphrase-multilingual-MiniLM-L12-v2, Table Transformer Detection, and Bielik-11B-v3.0-Instruct are now available in Amazon SageMaker JumpStart
Published: April 5, 2026
AWS has added three powerful models to Amazon SageMaker JumpStart: paraphrase-multilingual-MiniLM-L12-v2, Microsoft Table Transformer Detection, and Bielik-11B-v3.0-Instruct.
Paraphrase-multilingual-MiniLM-L12-v2 (Sentence Transformers): Lightweight semantic similarity model mapping sentences/paragraphs to 384D vectors in 50+ languages. Great for cross-lingual search, clustering, and similarity—zero language tweaks needed.
Microsoft Table Transformer Detection: DETR-based detector trained on PubTables-1M for spotting tables in PDFs/images. Ideal for digitizing docs and extracting data from papers or reports at scale.
Bielik-11B-v3.0-Instruct (SpeakLeash/ACK Cyfronet AGH): 11B-param generative model for 32 European languages (Polish focus). Shines in dialogue, STEM/math, logic, tools, and enterprise linguistic tasks.
Deploy with a few clicks in SageMaker Studio or via Python SDK. Full docs in Amazon SageMaker JumpStart.
9. AWS Deadline Cloud now supports custom scripting for job submission workflows
Published: April 2, 2026
AWS Deadline Cloud now lets you run custom scripts before and after job submissions, seamlessly weaving your pipeline into the process. This managed service streamlines render management for VFX, films, TV, web, and design teams.
Pre-submission scripts validate configs, add files (e.g., textures), tweak params, or enforce policies—before uploads. Post-submission scripts handle notifications, tracking, or logging. Define in YAML/JSON in your job bundle or shared dir (via env var). Scripts get job metadata, with timeouts.
Dive in via Deadline Cloud Client documentation.
10. AWS Transfer Family Terraform module now supports Okta and Microsoft Entra ID integration examples
Published: April 1, 2026
The AWS Transfer Family Terraform module now features complete examples for Okta and Microsoft Entra ID as custom identity providers (IdP). Automate secure file transfers tied to your existing identity setup.
Built on the open-source Custom IdP solution, it includes MFA, audit logs, and IP allowlisting. Okta covers password auth, TOTP MFA, and attributes; Entra ID handles password flows for Microsoft shops.
Grab it from the Terraform Registry. More in the user guide and AWS Capabilities table.
Summary
Key Highlights
- Performance Improvements: Enhanced scalability and efficiency across multiple AWS services
- Security Enhancements: New features to strengthen cloud security posture
- Developer Experience: Improved tools and integrations for faster development
- Cost Optimization: New options for better resource management and cost control
What This Means for You
These updates showcase AWS’s dedication to:
- Boosting app performance and reliability
- Bolstering security and compliance
- Simplifying operations
- Speeding up cloud app launches
Conclusion
April 2026’s AWS updates deliver game-changing advances in infrastructure, security, and dev tools. Whether building fresh apps or refining current ones, these features unlock smarter cloud strategies.
Stay tuned for more, and visit the AWS What’s New page for the freshest announcements.
Last updated: May 2, 2026