Member: TysonPaul | Microsoft Community Hub
Retirement of Microsoft HPC Pack
Team Blog: Azure High Performance Computing (HPC)
Author: XinXin
Published: 08/25/2026
Summary: Microsoft HPC Pack, a Windows-based HPC scheduler, will be retired, with support ending August 27, 2027. No new updates, features, or enhancements will be provided after this date, and only limited security support is available during the retirement period. Customers are advised to migrate to Azure Batch or other Azure services for HPC workloads. Existing deployments will not be forcibly disabled but will be unsupported after the retirement date. Microsoft encourages early migration planning and feedback to facilitate a smooth transition and address customer concerns.
Azure DNS + Traffic Manager linked records
Team Blog: Azure Networking
Author: atiy
Published: 08/26/2026
Summary: Azure DNS Traffic Manager linked records allow direct, managed integration between Azure DNS and Traffic Manager profiles, enabling endpoint IP addresses to be returned without an intermediate CNAME. This improves DNS responses, supports zone-apex routing, and enhances DNSSEC compatibility while preserving Traffic Manager’s routing intelligence, health monitoring, and failover. The article provides a practical guide for configuring, validating, and testing this feature, currently in public preview, using a multi-region Contoso scenario. The feature simplifies DNS architecture and client experience, making global routing more seamless and secure, especially for root domains.
Azure Virtual Network routing appliance is now generally available
Team Blog: Azure Networking
Author: Anshu_Verma
Published: 08/04/2026
Summary: Azure Virtual Network routing appliance is now generally available, offering a managed, platform-native routing service for high-performance connectivity across Azure virtual networks. Designed for modern cloud demands—including AI, private connectivity, IPv6 adoption, and multi-region architectures—it enables scalable hub-and-spoke and private network models, simplifies operations, and supports both IPv4 and IPv6. With built-in resiliency, Azure-native management, configurable bandwidth, and integrated monitoring, it helps organizations build large-scale, efficient, and innovative network infrastructures, accelerating cloud and AI initiatives while reducing operational complexity.
Introducing the new Microsoft 365 Setup experience: A personalized dashboard for admins
Team Blog: FastTrack
Author: johnvincent
Published: 08/31/2026
Summary: Microsoft is launching a redesigned Microsoft 365 Setup experience for admins, featuring a personalized dashboard that centralizes deployment progress, onboarding, licensing, and readiness insights. The new setup offers tailored recommendations, direct product configuration, and easy access to guides and support, streamlining the deployment process. Admins can now spend less time searching and more time taking action, with integrated training, migration resources, and FastTrack support. This update aims to simplify Microsoft 365 deployments, making them more intuitive and efficient, with future enhancements planned based on admin feedback, including AI-powered insights and assistance.
Building a Fully Automated Azure Landing Zone Deployment Using Azure DevOps and Terraform
Team Blog: Azure Tools
Author: Rajasekharvemula
Published: 08/18/2026
Summary: The article explains how to build a fully automated Azure Landing Zone using Azure DevOps and Terraform. It outlines best practices for deploying standardized cloud environments, including structured Git workflows, automated CI pipelines, environment promotion strategies, governance integration, identity and access management, and monitoring. Storing Terraform state in Azure Storage ensures collaboration and recovery. Automation from the start enables consistent, secure, and auditable cloud foundations, reducing operational effort and improving compliance. Combining Azure DevOps and Terraform transforms cloud deployment into a scalable, governed, and enterprise-ready platform engineering practice.
Cost Management with Azure Resource Manager MCP
Team Blog: FinOps
Author: demiajayi
Published: 08/25/2026
Summary: Microsoft has announced new Cost Management features in Azure Resource Manager MCP, providing default tools for AI agents to integrate cost insights into Azure workflows. These tools help with cost estimation, budget tracking, and identifying savings opportunities, while optional advanced tools enable deeper analysis, forecasting, and optimization. AI agents can query costs, analyze AKS workloads, and recommend savings, streamlining financial decision-making in cloud operations. Installation and configuration details are provided for VS Code and GitHub Copilot. Microsoft plans to expand agentic cost capabilities and invites user feedback on the new tools.
Who Ordered All These Tokens? Giving AI Spend a Name in FOCUS
Team Blog: FinOps
Author: Dirk_Brinkmann
Published: 08/19/2026
Summary: The article discusses the challenge of accurately attributing AI-related costs, such as Copilot and agent spend, within enterprise billing systems. Existing cost reports scatter AI expenses across ambiguous categories, making analysis difficult. The proposed solution is to create a custom “meter map” that classifies each billing meter into meaningful attributes (like token type, provider, Copilot, or agent). This enables clear reporting and chargeback, aligning spend with business questions. The approach is flexible, ongoing, and requires ownership, but solves the gap until vendor-neutral standards evolve further.
Zonal Resiliency in Azure: Application-Centric Goals, Recovery Plans, and Drills
Team Blog: ITOps Talk
Author: Pierre_Roman
Published: 08/06/2026
Summary: The article introduces “Resiliency in Azure” (formerly Azure Business Continuity Center), a unified platform for managing and validating zonal resiliency at the application level across IaaS, PaaS, and shared services. Key features include application-centric recovery plans, tailored Azure Advisor recommendations, Copilot-guided remediation, and automated zone-down drills via Azure Chaos Studio. This approach streamlines resiliency management, supports compliance, and enables realistic failover testing, reducing manual effort and uncertainty. While not all services are supported and costs may vary, it offers a significant improvement in visibility, confidence, and operational readiness for Azure applications.
Operating Azure Backup at Scale: Day-2 Excellence for IaaS, PaaS, and Storage Workloads
Team Blog: ITOps Talk
Author: Pierre_Roman
Published: 08/11/2026
Summary: The article summarizes advancements in Azure Backup, highlighting its unified, application-centric approach for protecting IaaS, PaaS, AKS, PostgreSQL, and storage workloads. Key features include a single vault model, built-in cyber resiliency with immutability and multi-user authorization, threat detection, agentless backups, and conversational configuration via VS Code. Azure Backup offers granular restores, compliance-friendly retention tiers, and robust protection against ransomware. Real-world scenarios and actionable steps demonstrate improved operational efficiency, security, and reliability for IT professionals managing large-scale Azure environments.
Expanding the Azure Local Security Ecosystem with IBM QRadar and Splunk
Team Blog: Azure Arc
Author: Ariel Netz
Published: 08/07/2026
Summary: Microsoft has announced new integrations for Azure Local with IBM QRadar and Splunk, enabling organizations in regulated and sovereign environments to incorporate Azure Local security signals into their existing security operations platforms. This enhances monitoring, investigation, and compliance while preserving established tools and processes. Azure Local’s built-in security controls and audit logs now work seamlessly with these partner solutions, supporting flexible and resilient security operations. Microsoft is also working to expand its security ecosystem with additional partners in areas like endpoint detection, vulnerability management, and container security for Azure Local and Sovereign Private Cloud deployments.
Episode 1: Onboarding Azure Arc at Scale | The Azure Arc Check-In
Team Blog: Azure Arc
Author: Mason_Torres
Published: 08/18/2026
Summary: Episode 1 of The Azure Arc Check-In discusses strategies for onboarding large fleets of Windows and Linux servers to Azure Arc at scale. It emphasizes automation and non-interactive authentication to minimize manual effort and errors. For Linux servers, Ansible with managed identity streamlines agent deployment, while Windows servers can be onboarded using Active Directory Group Policy. The episode provides step-by-step guidance for both approaches, highlighting the importance of aligning onboarding methods with existing management tools and enabling consistent configuration and governance from day one.
Announcing General Availability of Disable Multithreading and Configurable Constrained Cores
Team Blog: Azure Compute
Author: misha-bansal
Published: 08/19/2026
Summary: Microsoft Azure has launched two new features: Disable Simultaneous Multithreading/Hyper-Threading (SMT/HT Off) and Configurable Constrained Cores. These allow customers to optimize VM performance and reduce software licensing costs by customizing vCPU configurations and disabling SMT/HT for improved performance and latency. Users can select specific vCPU counts without altering memory or storage, aiding compliance and cost efficiency. The features are available in all Azure regions, with no changes to VM pricing, and help customers tailor compute resources to their workloads, maintaining performance while controlling costs.
The OpenVMM Project
Team Blog: Windows OS Platform
Author: Caroline_Perezvargas
Published: 08/17/2026
Summary: OpenVMM is an open-source, cross-platform Virtual Machine Manager (VMM) written in Rust and released under the MIT license. It offers broad support for multiple architectures, host operating systems, and virtualization backends, and features a modular, flexible design. Since becoming open source in 2024, OpenVMM has grown with contributions from major industry partners, expanding device emulation, testing infrastructure, and confidential computing capabilities. Its modern security architecture leverages Rust, process sandboxing, and user-mode virtualization. OpenVMM is used in various environments, including Azure, and aims to provide secure, high-performance virtualization for diverse workloads.
Azure Managed Redis Deployment Automation
Team Blog: Azure PaaS
Author: LuisFilipe
Published: 08/21/2026
Summary: The article introduces a PowerShell script for automating Azure Managed Redis (AMR) deployments, especially when facing regional capacity restrictions. The script retries cache creation at random intervals, cycling through specified SKUs and configurations, and cleans up failed resources before each attempt. It supports geo-replication, clustering, high availability, and custom eviction policies. Authentication, logging, and parameter validation are included. The script is provided as-is, without support or guarantees, and users are advised to test before production use due to potential costs and risks.
Tracking Batch node state and duration in Log Analytics
Team Blog: Azure PaaS
Author: Ahmed_Khaled
Published: 08/25/2026
Summary: The article addresses gaps in Azure Batch node monitoring, such as identifying nodes stuck in certain states and automating alerts. Since Azure Monitor lacks per-node granularity, the proposed solution uses a Python script to poll Batch REST API and ingest node-level data into Log Analytics. Custom KQL queries and alerts enable precise tracking of node state, duration, and errors. The solution supports dashboards, automated remediation, and actionable alerts, and can be deployed via Azure Functions, Container Apps, or VMs. It enhances monitoring without replacing built-in metrics, providing detailed, per-node visibility for customers.
Beyond Deployment: What It Really Takes to Run GitHub Actions Runners on AKS
Team Blog: Azure Infrastructure
Author: aparnabadireddigari
Published: 08/17/2026
Summary: Deploying GitHub Actions runners on Azure Kubernetes Service (AKS) with Actions Runner Controller (ARC) is straightforward, but running it as a production platform requires much more. Key challenges involve managing authentication, network dependencies, workload isolation, image lifecycle, and observability. Production readiness demands clear ownership, validated connectivity, strong operational practices, and reliable monitoring. Naming conventions and image management become platform contracts. Ultimately, success depends on operational excellence, governance, and ownership—moving beyond a working deployment to a mature, dependable platform that reliably supports multiple teams and business-critical workflows.
Azure Copilot Introduces Direct Access to Agents
Team Blog: Azure Infrastructure
Author: artigulwadi
Published: 08/18/2026
Summary: Azure Copilot now offers direct access to specialized agents for troubleshooting, deployment, optimization, and resiliency, allowing cloud operations teams to quickly engage relevant expertise. This update streamlines workflows by matching tasks with purpose-built agents, enhancing efficiency and tailored guidance. Administrators can manage agent availability flexibly, enabling incremental adoption and better governance. Pricing remains unchanged, with most agents available at no extra cost. This evolution supports agentic cloud operations, empowering teams to operate more effectively across application and infrastructure lifecycles, with future enhancements and administrative controls expected.
Accelerate Inference on AKS with Azure Blob Storage and NVIDIA Dynamo
Team Blog: Azure Storage
Author: Vishnu Charan TJ
Published: 08/11/2026
Summary: The article details how integrating Azure Blob Storage with NVIDIA’s Dynamo stack accelerates AI inference on Azure Kubernetes Service (AKS). By using Dynamo ModelExpress and Run:ai Model Streamer, model weights are loaded up to 7.3x faster, reducing cold start latency by up to 4.6x. Offloading KV cache via NIXL to Blob Storage further lowers time-to-first-token latency by 2.8x and optimizes GPU memory use. These enhancements enable faster scaling, operational simplicity, and improved reliability for large AI models, streamlining inference workflows on AKS.
Unified AI Defense: Security Copilot, Project Perception, and MDASH
Team Blog: Core Infrastructure and Security
Author: edgarus71
Published: 08/24/2026
Summary: The article outlines Microsoft’s unified AI security platform, comprising Security Copilot, Project Perception, and MDASH. Security Copilot assists security analysts via AI-driven workflows and investigations. Project Perception orchestrates multi-agent defense, coordinating Red, Blue, and Green agents to proactively identify, validate, and remediate threats. MDASH specializes in code vulnerability discovery, validation, and remediation guidance, feeding findings into broader defense workflows. Each tool serves distinct roles—analyst assistance, coordinated defense, and deep application security—but together provide a comprehensive, layered AI security solution with human oversight and governance.
AI Skills Navigator is now available in Microsoft Copilot
Team Blog: Microsoft Learn
Author: MattErni
Published: 08/27/2026
Summary: Microsoft has integrated AI Skills Navigator into Copilot’s Learning Agent, enabling users to ask questions and receive personalized AI skill recommendations directly in their workflow. This new feature tailors training, learning paths, and credentials to individual roles and goals, streamlining skill development for both individuals and organizations. Progress is tracked automatically, and learning resources from various platforms are accessible within Copilot. This integration helps connect real work tasks to relevant learning opportunities, making AI upskilling more efficient and accessible across organizations.
From GPO to Microsoft Intune: A practical guide to cloud-first policy management
Team Blog: Intune Customer Success
Author: Intune_Support_Team
Published: 08/31/2026
Summary: The article offers a practical guide for organizations transitioning from Group Policy Objects (GPOs) to Microsoft Intune for cloud-first policy management. It outlines three migration paths—starting fresh for cloud-native devices, selectively transitioning necessary settings, or coordinating GPO and Intune for hybrid environments. Key recommendations include assessing and rationalizing existing policies, avoiding direct lift-and-shift, leveraging security baselines, piloting changes, and coordinating targeting to prevent conflicts. The guide stresses the importance of tailoring management strategies to device populations and business needs, ensuring a streamlined, secure, and efficient move to modern policy management.
Introducing device association for Windows Autopilot device preparation
Team Blog: Intune Customer Success
Author: Intune_Support_Team
Published: 08/27/2026
Summary: Device association for Windows Autopilot device preparation enables organizations to securely bind Windows 11 devices to their tenant before enrollment, using hardware-backed attestation. This ensures devices are recognized as corporate-owned, allowing IT admins to apply targeted policies and configurations early, streamlining the out-of-box experience for users. The association persists through resets and reinstallation, supporting consistent lifecycle management. Device association integrates with existing Autopilot strategies, offering improved security, predictability, and a device-centric onboarding process, while simplifying setup for employees and enhancing control for IT teams.
From Features to Flow: How Real-World Adoption Reshaped the Azure Architecture Diagram Builder
Team Blog: Azure Architecture
Author: arturoqu
Published: 08/13/2026
Summary: The article discusses how real-world adoption of the open-source Azure Architecture Diagram Builder led to a shift from feature-focused development to a guided, four-stage workflow: Create, Refine, Validate & Improve, and Share or Build. User feedback revealed the importance of preserving human edits, improving diagram integrity, and ensuring validation continuity. Adoption data showed users required more than simple diagram generation—they needed iterative refinement, validation, and artifact export. The product now emphasizes workflow coherence, human judgment, and clear boundaries between logical and physical architecture, while maintaining privacy-safe feedback and ongoing open-source development.
Choosing the Right Agent in Microsoft Foundry
Team Blog: Azure Architecture
Author: supriyas
Published: 08/19/2026
Summary: The article discusses how to choose between Prompt Agents and Hosted Agents in Microsoft Foundry Agent Service. Prompt Agents are simpler, managed by Foundry, and suitable for straightforward tasks with clear instructions. Hosted Agents offer greater control, supporting complex workflows, custom code, and orchestration logic, but require more engineering responsibility. The key decision factor is runtime control—teams should start with Prompt Agents and switch to Hosted Agents only when customization or advanced orchestration is needed. Identifying runtime requirements clarifies which agent type is appropriate, minimizing risk and operational overhead.
[Now Generally Available] What-If for Azure Deployment Stacks
Team Blog: Azure Governance and Management
Author: torreyt
Published: 08/18/2026
Summary: Azure’s Stacks What-if feature is now generally available, providing detailed previews of changes before deploying Azure Deployment Stacks. It shows resource-level actions (create, modify, delete, detach), filters out false positives, and stores results as retrievable resources for use in CI pipelines. This enables safer updates by clarifying impacts, especially deletions or detachments, and supports review and approval workflows. Available across all regions and scopes via CLI, PowerShell, and SDKs, Stacks What-if addresses a top user request by enhancing deployment stack transparency and reliability.