Announcing the public preview of the new Azure FXv2-series Virtual Machines
Today, Microsoft is announcing the public preview of the new Azure FXv2-series Virtual Machines (VMs), based on the 5th Generation Intel® Xeon® Platinum 8573C (Emerald Rapids) processor. The preview includes the Compute-optimized VMs FXmsv2-series and FXmdsv2-series.
To request access to the preview, please fill out FXv2-series-Preview-Signup.
The new Azure FXv2-series VMs offer several enhancements compared to previous generation (FXv1-series) VMs:
- up to 1.5x CPU performance
- 2x vCPUs, with 96 vCPU as the largest VM size
- 1.5x+ Network bandwidth, and offers up to 70Gbps
- up to 2x local storage (Read) IOPS and offers up to 5280 GiB local SSD capacity
- up to 2x IOPS and up to 5x throughput in remote storage performance
- up to 400k IOPS and up to 11 GBps throughput with Premium v2/ Ultra Disk support
- up to 1800 GiB memory
FXv2-series VMs feature an all-core-turbo frequency up to 4.0 GHz. In addition to the standard sizes, FXv2-series VMs are available in constrained-core sizes, with vCPU count constrained to one-half or one-quarter of the original VM size, giving you the flexibility to select the core and memory configuration that best fits your workloads. These VMs support up to a 21:1 memory-to-vCPU ratio with the base sizes, and an even better ratio with the constrained-core sizes. You also have the option to choose between FXv2-series VMs with and without local disk. These VMs support NVMe interface for both local and remote disks.
Additionally, improvements in Azure Boost enhance the networking, storage and overall CPU performance and security of FXv2-series VMs. Finally, these VMs have enhanced AI capabilities with Intel® Advanced Matrix Extension (AMX), delivering higher inference and training performance, and enhanced security capabilities with Total Memory Encryption (TME) technology.
You can learn more about the new Azure FXv2-series VMs by visiting the specification pages at FXmdsv2-series and FXmsv2-series.
Overall, the FXv2-series Azure Virtual Machine is best-suited to provide a balanced solution for compute-intensive workloads such as databases, data analytics workloads and EDA workloads, that also require large amounts of memory and high-performance, storage, I/O bandwidth.
The FXv2-series is purpose-built, designed to address several requirements of SQL Server workloads, particularly those that require higher computing performance, memory, storage, I/O bandwidth but not necessarily a higher core count. Key benefits for SQL Server workloads include:
- Improved performance: The FXv2-series VMs offer higher and improved performance capabilities, ensuring SQL Server workloads can handle large amounts of data and transactions efficiently. These capabilities are crucial for applications that require high I/O operations per second (IOPS) and low latency.
- High memory and storage: FXv2-series VMs provide specifications (such as 96 vCPU with up to 1800 GiB memory, up to 400K IOPS and up to 11 GiB remote storage throughput) to cater to the needs of SQL workloads requiring high memory, storage, and I/O bandwidth.
- Cost efficiency: Constrained vCPU capable VM sizes help reduce the cost of SQL Server licensing by offering an improved memory-to-core ratio while maintaining the same memory, storage, and I/O bandwidth.
- Local storage: FXmdsv2-series VMs provide local storage for tempdb heavy workloads, significantly improving the performance of SQL Server workloads that need quick access to temporary data.
FXv2-series VMs also provide EDA customers with an updated Intel-based option tailored to the needs of the semiconductor community. Features like larger L3 caches, higher instructions per clock, spacious memory, and faster storage compared to the prior generation FXv1 VM combined to make FXv2-series VMs much improved for the demanding needs of EDA workloads, such as reducing chip design turn-around time and time-to-market and improving license utilization.
During the preview period, the VMs are available in Azure regions such as West US 3 and Southeast Asia for all VM sizes up to 96 vCPUs. The number of regions will continue to expand in 2024 and beyond.
The new virtual machines support all remote disk types such as Standard SSD, Standard HDD, Premium SSD (v1), Premium SSD v2 and Ultra Disk storage. To learn more about various disk types and their regional availability, please refer to Azure managed disk type and disk storage is billed separately from virtual machines. You can deploy these new VMs using existing methods including Azure portal, SDKs, APIs, PowerShell, and the command-line interface (CLI).
The new Azure FXv2-series VMs support a wide range of OS distributions, such as Canonical Ubuntu, Red Hat Enterprise Linux, SUSE Enterprise Linux, Alma Linux, Windows Client 11 Pro and Enterprise, Windows Server and many more, for developers who want to take advantage of Azure’s highly available, scalable, and secure platform to run cloud-based workloads and test workflows. Customers can access the full list of images in the Azure Marketplace.
During the preview, Azure FXv2-series VMs will be charged at full price. To learn more about the pricing of FXv2-series VMs, please visit the Azure Virtual Machines pricing pages. Eligible new Azure customers can sign up for an Azure free account and receive $200 Azure credit.
FXv2-series VMs are tailored for compute-intensive tasks, making them perfect for high-performance data processing, complex operations, and intensive analytics, delivering excellent performance and efficiency. The new FXv2-series VMs provide a versatile solution for a wide range of computing needs. Explore the FXv2-series VMs today and discover how these VMs can enhance your performance and operational efficiency.
To request access: please fill out FXv2-series-Preview-Signup.
Have questions ? Please reach us at Azure Support and our experts will be there to help you with your Azure journey.
Published on:
Learn moreRelated posts
Upcoming Changes to Instance Size Flexibility Ratios for Reserved VM Instances for M-series: What Yo
Overview In the ever-evolving landscape of cloud computing, it is crucial to stay informed about changes that may affect your usage and busine...
Exploring SUSE Enterprise Linux on Azure
Exploring SUSE Enterprise Linux on Azure In today's cloud-centric world, leveraging robust and reliable operating systems is crucial for busin...
Announcing Public Preview of new attach/detach disks API for VMs/VMSS
We are excited to announce the public preview of a new API that will make attaching and detaching disks to a VM faster and easier. The new API...
Fine-tuning a Hugging Face Diffusion Model on CycleCloud Workspace for Slurm
Introduction Azure CycleCloud Workspace for Slurm (CCWS) is a new Infrastructure as a Service (IaaS) solution which allows the users to purpos...
Public Preview Announcement-On Demand Capacity Reservation in Azure in China
Today, we're announcing the public preview of on demand capacity reservations for Azure Virtual Machines in Azure in China Cloud . Y...
Breaking change for Window Server 2022 Image Users with .NET 6
Azure Marketplace media images for Windows Server 2022 currently include .NET 6, but going forward will not include a .NET version with the im...
Effortlessly Migrate Azure VMs between zones
For various reasons you might come across a situation when you need to migrate your Azure VMs from one zone to another. Migrating Azure VMs be...
Announcing Public Preview of Instance Mix on Virtual Machine Scale Sets
Today, we’re excited to announce that the ability to specify multiple different VM sizes in your Virtual Machine Scale Sets (VMSS) with Flexib...
Announcing General Availability of Attach & Detach of Virtual Machines on Virtual Machine Scale Sets
Today, we’re thrilled to announce that the ability to attach or detach Virtual Machines (VMs) to and from a Virtual Machine Scale Set (VMSS) w...
Recovery options for Azure Virtual Machines (VM) affected by CrowdStrike Falcon agent
We are aware of an issue that started on 19 July 2024 at 04:09UTC, which resulted in customers experiencing unresponsiveness and startup failu...