Oct 21 - Oct 27 Updates

Last Week Cloud Platform Updates

October 30, 2024

Oct 21 - Oct 27 | AWS, Azure & GCP Updates

AWS now allows customers to make partial payments towards their monthly bill using credit or debit cards. Instead of paying the full amount at once, customers can split their payments across multiple cards, directly from the AWS Billing Console. This feature eliminates the need to contact AWS Customer Service and provides flexibility for managing payments. To make a partial payment, simply select an invoice, choose a card, edit the amount, and confirm.

Amazon Elastic Kubernetes Service (EKS) now integrates with Amazon Application Recovery Controller (ARC), supporting zonal shift and zonal autoshift to enhance application availability. These features allow EKS users to quickly redirect network traffic away from impaired Availability Zones (AZs), ensuring high availability for multi-AZ Kubernetes applications. Zonal shift enables manual traffic redirection, while zonal autoshift automates the process, allowing for practice runs to test cluster resilience with one less AZ.

Amazon EKS now offers dual stack support for its management API and Kubernetes API server endpoints in IPv6 EKS clusters, allowing connections over IPv6, IPv4, or dual stack clients. Dual stack is also supported for private access via AWS PrivateLink, using a new AWS DNS domain name while maintaining existing endpoints for backward compatibility. This update facilitates gradual migration from IPv4 to IPv6, helping meet IPv6 compliance and eliminating the need for expensive address translation equipment. Note, this change applies to EKS managed endpoints and does not affect pod networking.

AWS has introduced enhanced monitoring for Amazon ECS applications via Amazon CloudWatch Application Signals, an APM feature that tracks application performance against key business or service level objectives (SLOs). This update adds infrastructure metrics correlation to existing traces and logs, providing a comprehensive view of application health without manual effort or custom code. By correlating application metrics, traces, logs, and infrastructure data, Application Signals helps speed up troubleshooting and reduce disruptions. For example, developers can investigate latency issues and correlate them with ECS task metrics, such as CPU or memory usage, for quicker resolution.

AWS Lambda has introduced a new code editor in the Lambda console, powered by Code-OSS (VS Code – Open Source). This update provides an intuitive and familiar coding environment, offering features like the command palette, quick search, and customizable themes. The editor enhances productivity with real-time code suggestions via the Amazon Q Developer extension and allows developers to view both function code and test results simultaneously, streamlining the serverless application development and debugging process.

Enterprise and Enterprise Flash tier caches now have the ability to scale up or out without requiring downtime. Scaling up allows a cache to scale to a larger Enterprise SKU with more memory capacity, connections, and compute performance. Scaling out spreads data over additional nodes, which likewise boosts memory and compute performance. With in-place scaling, the scale operation occurs without disrupting the operation of the current cache, allowing you to use the same cache resource with minimal interruption and no need to change cache addresses.

You can now set fail criteria on server metrics, thereby defining performance and quality expectations for your application under load. Monitor your applications components and specify the thresholds on server metrics for test failure.

Azure Functions now supports Node.js 22 in preview. You can now develop functions using Node.js 22 locally and deploy them to all Azure Functions plans on Linux and Windows.

The Premium tier for Azure App Configuration is now generally available, making it possible to support larger and more complex applications with higher performance and scalability.

This new pricing plan is ideal for organizations that need greater capacity, enhanced security, and global availability for configuration management.

  • Increased Capacity: Supports higher configuration volumes and increased request throughput for large-scale applications.

  • Included Global Replication: 1 included replica ensures faster access and increased resilience through replication across multiple regions.

  • Improved SLA: Offers 99.99% SLA, tailored for mission-critical applications.

Cloud Logging and Cloud Monitoring now support creating SQL-based alerting policies to monitor the results of your SQL queries. This allows for more targeted monitoring and incident management based on specific query outcomes. For more details, see the documentation on monitoring SQL query results and managing incidents for SQL-based alerting policies.

Connecting to Cloud Storage using gRPC is now generally available (GA), enabling efficient interactions with Cloud Storage. Additionally, you can now emit client-side metrics for gRPC to monitor and optimize performance. For more details on supported metrics, see the documentation on using gRPC client-side metrics.

  • Backup Charges: The maxChargeableBytes parameter now appears in the backupRuns.GET API and gcloud sql backups describe, indicating the maximum chargeable backup size.

  • PostgreSQL 17 General Availability: PostgreSQL 17 is available, with Enterprise Plus as the default edition for version 16 and later. Several flags and extensions are deprecated, including old_snapshot_threshold and oracle_fdw.

  • Read Replicas: Read replicas can now be created for instances with private services access and connector enforcement enabled across all three databases.