VM vCore customization features disabling simultaneous multi-threading

Public Preview: VM vCore customization features disabling simultaneous multi-threading (SMT/HT) and constrained cores

Azure announces the public preview of VM Customization features, a new capability that gives customers greater control over virtual CPU configurations to optimize performance and licensing costs. This feature introduces two powerful options: Disable Simultaneous Multi-Threading (SMT/HT) and Configurable Constrained Cores.

With VM Customization, customers can now:

• Disable Simultaneous Multithreading (SMT/HT) to run VMs with one thread per core for improved performance on latency-sensitive and single-threaded workloads
• Select a custom vCPU count from a guided list of valid options to reduce software licensing costs while retaining full memory, storage, and I/O bandwidth

These capabilities are available across a wide range of VM sizes and are supported in select regions during public preview.

VM Customization offers:

• Self-service control over Simultaneous Multithreading (SMT/HT)
• Flexible vCPU selection from validated core count options per VM size
• Support for both features independently or in combination
• Reduced licensing costs for software billed per vCPU (e.g., SQL Server, Oracle)
• Azure Portal, ARM template, CLI and PowerShell support for easy deployment

These features are ideal for database workloads, HPC scenarios, and customers seeking to optimize licensing costs without sacrificing performance.

Generally Available: Azure SQL updates for mid-October 2025
In mid-October 2025, the following updates and enhancements were made to Azure SQL:
Improve the redirect connection type to require only port 1433 and promote to default. Learn more.

Convert Azure SQL Database to Hyperscale while preserving existing geo-replication or failover group settings.

Generally Available: Near-zero downtime scaling for HA-enabled Azure Database for PostgreSQL servers
You can now scale your Azure Database for PostgreSQL servers that have high availability (HA) enabled with minimal impact to your users. The process usually takes less than 30 seconds, while traditional scaling can take 2 to 10 minutes. This new capability ensures near-zero downtime during scaling operations, allowing you to maintain business continuity while adapting to changing workloads. Whether you're responding to seasonal demand spikes or expanding your application footprint, this enhancement gives you the flexibility to scale seamlessly.

Open Source: Containerization Assist MCP Server
Containerization Assist solves the challenge of manual, error-prone containerization by automating Dockerfile creation and Kubernetes manifest generation. Unlike basic AI coding tools, it is a full containerization platform built on AKS Draft’s proven real-world foundation. It uses LLM-powered artifact generation with iterative feedback from local builds and deployments to improve accuracy and efficiency. Developers can integrate it into IDEs like VS Code with MCP server support, streamlining workflows for modernizing legacy applications

Generally Available: Cloud-to-Cloud migration made simple with Azure Storage Mover
Azure Storage Mover’s new AWS S3 to Azure Blob migration feature is now generally available, empowering organizations to move data directly from AWS to Azure securely, reliably, and at scale. With this release, customers can eliminate manual pipelines and third-party tools, automate migrations in the Azure portal, and gain real-time visibility into every job.

Key benefits:

1. Fully Managed and Scalable

Azure Storage Mover is a fully managed service, meaning Microsoft handles the infrastructure, scaling, and reliability. This reduces operational overhead and ensures consistent performance across large-scale transfers.

1. No On-Premises Agent Required

Unlike on-premises migrations, cloud-to-cloud transfers do not require deploying a self-hosted agent. This simplifies setup and reduces the need for additional compute resources.

1. High-Speed, Parallel Transfers

Storage Mover supports parallel file transfers, enabling faster migration of large datasets. This is particularly useful when moving data across regions or between storage accounts with high throughput requirements.

Get started today and simplify your multicloud data estate with Azure Storage Mover’s Cloud-to-Cloud migration.

Public Preview: Sharing Capacity Reservation Groups
The ability to share Capacity Reservation Groups (CRGs) with subscriptions is now in Public Preview.
Previously, customers could deploy VMs in a Capacity Reservation Group within the same subscription only. On-demand Capacity Reservation Group (CRG) can now be shared with other subscriptions. Using this option can make it easier to manage some common configuration needs like resource reuse, centralized capacity management, more cost-effective scale-out and separating security and capacity concerns.

Generally Available: Enhanced cloning and Public IP retention scripts for Azure Application Gateway migration
Azure Application Gateway introduces two new PowerShell scripts to simplify migration from V1 (Standard or WAF) to V2 (Standard_V2 or WAF_V2) services.

As V1 SKUs’ is scheduled to retire by April 2026, we advise migrating early and take advantage of Application Gateway V2’s superior performance, scalability, monitoring, and unique features.

Enhanced cloning script:

• Automates configuration replication from an existing V1 gateway to a new V2 gateway.
• Eliminates manual entry of front-end TLS and trusted root certificates.
• Supports creation of private-only V2 gateways.

Public IP retention script:

Enables retention of existing V1 public IP address with V2 gateway.

Generally Available: New OS SKU enum to migrate to Azure Linux 3.0
You can now migrate to Azure Linux 3.0 with a new OS SKU enum, ‘AzureLinux3’, in Kubernetes version 1.28-1.36.

This new option provides increased control over OS upgrades by decoupling them from Kubernetes upgrades. Customers using LTS version 1.28-1.31 can now migrate from Azure Linux 2.0 to Azure Linux 3.0 without a Kubernetes upgrade.

Generally Available: Azure Functions support for Python 3.13
You can now develop functions using Python 3.13 locally and deploy them to Azure Functions.

Starting with Python 3.13, Functions introduces runtime version control, a new opt-in feature that lets you target specific versions of the Functions Python runtime used by your app. Python 3.13 also introduces several enhancements to Functions that improve performance and reliability. In Python 3.13+, Azure Functions no longer supports custom worker extensions and shared memory for large payloads, requiring migration to supported alternatives and gRPC-based messaging.

Generally Available: Prescaling in Azure Firewall
Azure Firewall now supports prescaling, a feature that enables admins to provision and reserve capacity units in advance to handle expected traffic loads – such as seasonal peaks or planned business events. This helps to ensure consistent throughput, faster scaling response, and greater control over capacity.

Key highlights:

• Proactive planning: Pre-allocate capacity to meet demand spikes and critical workloads
• New observability metric: View current and historical capacity usage in the Observed Capacity Metric to provide strategic insight into planning
• Flexible billing: Pay only for the provisioned capacity units, with the flexibility to adjust as your needs evolve.

Prescaling is available for Azure Firewall Standard and Premium SKUs in all public regions.

Generally Available: Observed capacity metric in Azure Firewall
Azure Firewall now includes the observed capacity metric — a new observability signal designed to help administrators understand how their firewalls are scaling in practice. Observed capacity tracks the number of capacity units actively utilized over time, offering valuable insight into real-world scaling patterns and performance behavior.

Key highlights:

• Validate setup: You can validate your prescaling or autoscaling setup is performing as expected.
• Proactive alerting: Set alerts when usage near defined thresholds and diagnose whether scaling keeps paced with demand
• Forecasting: Plan future capacity needs using historical and current traffic trends

Generally Available: PowerShell, Az cmdlets, and Python SDK for Azure Database Migration Service
Announcing the general availability (GA) of automation support for Azure Database Migration Service (DMS) through PowerShell, custom cmdlets, and the Python SDK.

Using Azure PowerShell Az.DataMigration or Azure CLI az datamigration, you can migrate databases by automating the creation of Azure DMS, configuring database migrations for online migration, and performing a cutover. For more commands and functionality, see Azure Samples.

This milestone unlocks powerful, stable, and scalable automation options for database migration workflows—making it easier than ever to integrate DMS into your DevOps pipelines and enterprise migration strategies.

Generally Available: Availability Zones and network restricted Key Vault and App Configuration references for Flex Consumption
You can now enable availability zones for your Azure Functions Flex Consumption apps during both creation and post-deployment. When enabled, instances are automatically distributed across zones, enhancing reliability and fault tolerance.

New Region Support: Availability zones are now also supported in East US and Southeast Asia.

Flex Consumption apps now support Key Vault and App Configuration references as app settings even when those resources are network restricted.

To use this feature ensure your app has outbound networking configured and confirm it can reach the referenced services.

Public Preview: Private Link Service Direct Connect
Azure Private Link service allows you to make your applications available to your customers privately and securely. The current setup procedure requires you to configure a private link service and place applications behind a standard load balancer. Private Link Service Direct Connect expands the functionality and allows you to directly connect a private link service to any routable destination IP address.

This feature is currently in limited public preview, available in the following regions with more to be added:

North Central US, East US 2, Central US, South Central US, West US, West US 2, West US 3, Asia Southeast, Australia East, Spain Central

Generally Available: SAP Business Data Cloud Connect to Azure Databricks
SAP Business Data Cloud (BDC) Connect to Azure Databricks is now generally available, enabling secure, bi-directional, zero-copy data sharing between SAP BDC and Azure Databricks using Delta Sharing, so teams can unify SAP and external data for analytics, BI, and AI without replication or governance gaps. This integration streamlines pipelines, reduces maintenance, and ensures SAP business context remains intact, powering advanced data products and AI on a trusted, governed foundation. Get started today in Azure Databricks to unlock live, contextual SAP data for your analytics and machine learning needs.

Generally Available: Azure Storage Discovery
Azure Storage Discovery provides enterprise-wide visibility into your Azure Storage data estate.

Insights that weren’t available before are now at your fingertips. Deeply analyze used capacity and activity, optimize costs, enhance security, and drive operational efficiency.

Azure Storage Discovery integrates with Azure Copilot. Unlock insights across your data estate and accelerate decision-making by simply asking for what you need to know. Whether you're a cloud architect, storage administrator, or data governance lead, Azure Storage Discovery helps you quickly answer key questions like:

• How much data is stored across all my storage accounts?
• Which regions are experiencing the fastest growth?
• Where can I reduce costs by adjusting access tiers or removing stale data?

Storage Discovery is available in two pricing plans:

• Free: Includes basic insights.
• Standard: Unlocks full insights.

Start analyzing your data estate across subscriptions within hours. We’ll even provide some historical data from before your deployment. Over time, Azure Storage Discovery enables you to analyze annual trends, like peaks and valleys in business workloads, with up to 18 months of data retention included.

Generally Available: Locations API Update for UK Azure Regions
To align with evolving compliance and regulatory requirements, Azure is updating the geographyGroup and regionalDisplayName metadata for UK-based regions in the Locations API.

These changes will take effect this month, October 2025, and will apply to the existing UK South and UK West regions. It will introduce a new geographyGroup value: "UK", and the regionalDisplayName prefix will also be updated from "Europe" to "UK".

What customers will see:

• Updated values in Locations API responses for UK-based regions
• Revised region names in Azure Portal region selection menus

Generally Available: Azure Event Grid new capabilities
Multiple new Azure Event Grid features have reached General Availability, supporting real-time telemetry, automation, and hybrid workloads.
Features now in General Availability:

MQTT OAuth 2.0 Authentication
Authenticate MQTT clients using JSON Web Tokens (JWTs) issued by any OpenID Connect (OIDC)-compliant identity provider, enabling seamless integration with Microsoft Entra ID (formerly Azure AD), custom identity platforms, or third-party IAM solutions.

MQTT Custom Webhook Authentication
Use external webhooks or Azure Functions to dynamically validate client connections. Authorize with shared access signatures (SAS), API keys, custom credentials, or X.509 certificate fingerprints to support scenarios requiring granular control across large, dynamic fleets of devices or multitenant environments.

MQTT Assigned Client Identifiers
Assign deterministic, pre-approved client identifiers to MQTT clients to offer session continuity, better device tracking, simplified diagnostics, and improved auditability—critical for managing long-lived connections and operational visibility in regulated industries.

First Class integration with Fabric
Route MQTT messages and Cloud Events from Event Grid Namespace to Fabric Event Streams for real-time analytics, storage, visualization of IoT data without using Event Hubs.

Generally Available: Spot Placement Score
Today, we are announcing the general availability of Azure Spot Placement Score, a feature that allows you to assess the likelihood of deployment success for your Spot virtual machines. Spot availability can vary across different combinations of Azure locations, availability zones, or VM sizes. Spot Placement Score provides the ability to evaluate the chance of deployment success on a near real-time basis, by assigning a rating of High, Medium, or Low for a given number of Spot VMs before deployment. This feature provides increased clarity and control, enabling you to make more informed decisions about which Spot VMs to deploy and where.
Using Spot Placement Score, you can achieve the following:

• A clear evaluation of how likely your Spot deployment is to succeed based on specified parameters.
• Identify the most suitable combination of regions and VM sizes to maximize Spot VM availability based on placement scores.
• Improve the overall success rate of deploying Spot VMs by leveraging data-driven placement scores, reducing the risk of capacity issues or failures during deployment.

Generally Available: Azure Site Recovery support for Ultra Disks
Announcing the General Availability of Azure Site Recovery (ASR) support for Virtual Machines with Ultra Disks. This release enables organizations to replicate, failover, and failback virtual machines with Ultra Disks across Azure regions with minimal impact on production performance. ASR offers automated recovery orchestration, cost-efficient replication, and non-disruptive testing, helping businesses ensure operational resilience, meet compliance needs, and minimize downtime. With this milestone, customers can now confidently extend enterprise-grade protection and continuity to their Ultra Disk–backed workloads using Azure Site Recovery.
Ultra Disks are the most performant block storage option for Azure virtual machines offering consistent low sub-millisecond latency, extremely high performance. This makes Ultra disk ideal for a broad spectrum of the most demanding enterprise production workloads, such as SAP HANA, top-tier databases, and transaction-heavy systems that require the highest levels of performance.