Microsoft Corp. said today it’s beefing up its Azure Kubernetes Service, with a slew of updates focused on enabling greater efficiency, cost optimization and scalability, plus improved networking and security and superior observability capabilities.

The updates were announced at this week’s KubeCon + CloudNativeCon NA event taking place in Chicago, the world’s largest annual Kubernetes conference. The open-source Kubernetes has emerged as one of the most critical enterprise software platforms of the modern age, used to manage container-based applications that can run anywhere.

Microsoft’s AKS is a cloud-based and fully managed implementation of Kubernetes that’s hosted on the Azure cloud platform. It has become one of the most popular Kubernetes distributions, used by thousands of enterprises to manage hundreds of modern applications.

Greater efficiency

In announcing today’s updates, Microsoft explained that because efficiency is a major concern for Kubernetes users in the current macroeconomic environment, it has put a lot of effort into making AKS more resource efficient. Those efforts are visible now with the general availability of kube-reserved resource optimization in AKS, which helps improve its resource reservation algorithms to deliver a reduction of up to 20% in required memory.

Further boosting efficiency, the company announced a public preview of Artifact streaming support in AKS. It enables customers to scale their workloads without waiting for container images to be fully loaded into Kubernetes clusters.

Improved scalability

In terms of scalability, Microsoft said it’s launching an Istio-based service mesh add-on for AKS in public preview, providing egress support and enabling customers to bring their own certificate authority and manage Istio upgrades more easily. A second add-on, for Kubernetes event-driven autoscaling, is now generally available to all customers, helping to make application autoscaling applications a much simpler process. It works by applying event-driven autoscaling to each application based on user demand, adding resources at busier times and reducing them when not required, the company said.

Further progress has been made in terms of storage flexibility too. The preview of Azure Container Storage in AKS means customers can now deploy dedicated container storage in 26 regions, a move that Microsoft said will simplify volume provisioning and management for stateful container applications. According to the company, this feature allows customers to more rapidly scale out of volumes, reduce pod failover times and lower the total cost of ownership for each application across multiple block storage options.

Networking and security updates

During Microsoft’s presentation, officials stressed that networking and image security remains a key challenge for companies using Kubernetes. To help ease these concerns, it’s previewing a new image integrity feature in AKS that can quickly verify container images originate from a trusted source and haven’t been tampered with.

It also announced dual-stack support for Azure CNI Overlay for AKS, available now in public preview. The update is said to enhance AKS’s networking capabilities, enabling both IPv4 and IPv6 addresses to co-exist within a single cluster, meaning users have more flexibility and control over connectivity.

There’s also a new application routing add-on for AKS, generally available now, which makes it easier for customers to get new web applications up and running. According to the company, it eliminates the need to set up an ingress controller, security certificate and perform domain name service management. In addition, it announced the public preview of Azure Backup for AKS, simplifying the process of backing up and restoring container applications and their data.

Enhanced observability and metrics

The final updates announced today are focused on improving observability for AKS workloads. With the expanded availability of Azure Monitor for Prometheus, Microsoft said more users now have a way to collect and analyze AKS metrics at scale using a Prometheus-compatible monitoring platform. With it, teams can use the Prometheus query language to analyze and receive alerts on the performance of monitored infrastructure with minimal complexity.

Finally, the company said customers can now take advantage of Azure Monitor OpenTelemetry-based distributions for Node.js and Python, in order to gain more visibility into failures, bottlenecks and usage patterns across Kubernetes clusters.

Image: Bing Image Creator