Weekly Azure Roundup: AKS tuning, Fabric pipelines, and IaC
Azure’s current updates offer improvements in networking, infrastructure automation, data engineering, operations, and developer productivity. These include enhanced reliability for Kubernetes and storage, automation features, better analytics, and tools for modern developer workflows.
Azure Kubernetes Service (AKS) and Infrastructure Automation
Guides for AKS cover how to scale, secure, and improve cluster performance. One walk-through investigates DNS scaling using Cilium, NodeLocal DNSCache, and FQDN policy to address latency in large workloads and documents troubleshooting for outbound traffic. Another resource explains optimizing AKS node pools with Crossplane, including notes on version compatibility and automation. Java users can use the Azure Performance Diagnostics Tool v5.0 to monitor JVM metrics—useful for faster debugging on Kubernetes.
- Scaling DNS on AKS with Cilium: NodeLocal DNSCache, LRP, and FQDN Policies
- Parallel AKS Node Pool Creation with Crossplane: A Version Compatibility Journey
- Automated Java Performance Diagnostics in Kubernetes using Azure SRE Agent
Microsoft Fabric Data Integration, Real-Time Intelligence, and Analytics
New features in Fabric include full pipelines for retail analytics using Delta Lake, Debezium, and Azure Event Hubs, showing automated change tracking and partition management. Data Factory now adds incremental copy/CDC, more connectors, flexible replication, and adjustable mapping for schema changes. A preview integration with Cribl Stream allows for fast telemetry routing and visualization.
- Scalable Data Ingestion for Retail: Dynamic Partitioning and Source Detection with Microsoft Fabric
- Enhancements to Microsoft Fabric Data Factory Copy Job: Incremental Copy and Change Data Capture
- Integrating Cribl with Microsoft Fabric Real-Time Intelligence (Preview)
Azure Storage: AI-Centric Platform Evolution
The Azure Storage service’s updated roadmap outlines shifts toward scalable, AI-focused workloads. This includes changes for blob storage, deep integration with AMLFS, and options for GPU-based operations (benefiting agents or LLMs). Elastic SAN and ACStor provide orchestration and combined file/block sharing for demanding deployments. Resiliency improvements span disks and files, while sustainability and smart tiering features are also being expanded.
Azure Automation, SRE, and Incident Management
Recent guides describe how to automate with Azure SRE Agents. You can connect SRE Agents with MCP for fine-grained roles and permission, scheduled compliance/security checks, and send automated reports via Teams or GitHub. Incident management integrates with ServiceNow to streamline triage and root cause processes, and work notes are now automated with AI summaries.
- How to Connect Azure SRE Agent to Azure MCP
- Proactive Cloud Ops with SRE Agent: Scheduled Checks for Azure Optimization
- Connect Azure SRE Agent to ServiceNow: End-to-End Incident Response
Azure Verified Modules, Infrastructure as Code, and Platform Foundations
Azure Verified Modules (AVM) for Platform Landing Zone with Bicep are now generally available, featuring modular IaC support for governance, network, and management. AVM adds Deployment Stacks, parameter files, policy control, and clear docs for migrations—continuing the platform update cycle.
Developer Tools, Testing Services, and Workflow Improvements
Developers now have access to the Azure Playwright Testing Service (Preview) for scalable UI/API automated tests. The preview includes workspaces, secret handling, CI integration, and reporting tools. Playwright Workspaces v2.0 add report views, artifact handling, and data retention controls to help with workflow governance and collaboration—building on ongoing improvements in SQL and CI pipelines.
- Running Playwright Tests at Scale with Azure Playwright Testing Service (Preview)
- Reporting Features Now Available in Playwright Workspaces on Azure
Memory Reliability and Hardware Efficiency for Azure Infrastructure
Azure launches RAIDDR, an open-source tool for improving reliability in modern memory (like LPDDR5X), and ELC for adaptive CPU power management, increasing data center energy savings while optimizing latency and performance. Both areas advance last week’s focus on infrastructure sustainability.
- RAIDDR: Redefining Memory Reliability for Hyperscale Azure Infrastructure
- Improving Efficiency through Adaptive CPU Uncore Power Management
Application and Container Networking, Security, and Cache Optimization
Guides for Azure Container Apps show secure integration with virtual networks and firewall routing for policy enforcement, monitoring, and compliance. Redis tips include scripts for listing key lifetimes, statistics, and tuning, assisting teams with troubleshooting and scaling.
- Advanced Container Apps Networking: VNet Integration and Centralized Firewall Traffic Logging
- Troubleshooting Azure Redis: Key TTL and Size Analysis with Bash and Lua
Azure Arc Server and Hybrid Cloud Updates
The Azure Arc Server recap covers improved management, zero-downtime patching, TPM rollout, and SQL hybrid workflows, maintaining last week’s focus on secure hybrid and multi-cloud management.
Microsoft Data Platform Ecosystem
SQLCon and FabCon announcements outline conference topics, training, and product updates for SQL Server, Azure SQL, Fabric, and AI-powered data management. These events extend previous coverage on data platform feedback and innovation.
Other Azure News
Developer-focused updates enhance debugging, performance, and workflow management. New filters for Azure Boards (now in private preview) let backlog and Kanban boards filter by custom fields, supporting better UI and management options.