By Zesty Team

Read More >

Kubernetes Updates – April

April 2025 brought a mix of exciting updates across The Big 3 – GCP, AWS, and Azure. From new features for AI workload optimization to improved performance, resource management, and monitoring capabilities, there’s a lot to unpack. 

While not all updates are game-changers, many provide valuable upgrades that can help teams optimize their Kubernetes clusters more effectively. Let’s dive into the highlights from this month!

Google Cloud

Automatic Application Monitoring – Generally Available in GKE 1.28+

GKE versions 1.28 and later now support Automatic Application Monitoring, which is generally available as of April 2025. This feature automatically collects key metrics using Google Cloud Managed Service for Prometheus and provides out-of-the-box dashboards for monitoring supported workloads. Additionally, it now supports six new AI model servers: NVIDIA Triton, vLLM, TGI, JetStream, TorchServe, and TensorFlow Serving.

GKE Inference Gateway – Generally Available
On April 17, 2025, Google announced the general availability of the GKE Inference Gateway. This feature enhances the performance, efficiency, and observability of generative AI workloads on GKE. With intelligent scaling and load balancing, it supports dense multi-workload serving of multiple LoRA fine-tuned models on a shared accelerator, leading to higher GPU/TPU utilization.

Workload Rightsizing Recommendations
GKE now provides insights and recommendations to help you identify workloads without resource requests or limits, allowing you to specify the resource needs for these workloads. Configuring CPU and memory requests and limits is a best practice for improving reliability and performance, and it’s necessary for understanding and optimizing resource utilization and associated costs.

Amazon Web Services

Bottlerocket Support for EKS Hybrid Nodes
On April 29, 2025, AWS announced the availability of Bottlerocket for EKS Hybrid Nodes. Bottlerocket is a Linux-based operating system optimized for running containers, providing improved security and performance. This support extends to EKS Hybrid Nodes, enabling secure containerized workloads across on-premises and AWS environments.

AWS Batch Now Supports Amazon ECS Exec and AWS FireLens Log Router
AWS Batch now supports Amazon ECS Exec and AWS FireLens log router for AWS Batch on Amazon ECS and AWS Fargate. With ECS Exec, you can track the progress of your application and troubleshoot issues by running interactive commands against the containers in your AWS Batch job. AWS FireLens allows you to stream logs of your AWS Batch jobs to various destinations, enhancing log management.

Microsoft Azure 

Azure Linux 3.0 Generally Available
Azure Linux 3.0 became generally available in April 2025, introducing enhanced performance, improved security, and better integration with Azure services. Kubernetes version 1.32 is the last version to support Azure Linux 2.0, with Azure Linux 3.0 being supported from Kubernetes version 1.31 onward.

Layer 7 Network Policies with Advanced Container Networking Services (ACNS)
Azure introduced Layer 7 network policies through its Advanced Container Networking Services offering. This allows for more granular control over egress traffic, enabling policy-driven management based on Fully Qualified Domain Names (FQDNs). The new functionality aims to bolster network observability and security within AKS clusters.