Azure just announced the public preview for the D and E family VMs powered by the new 5th Gen Intel® Xeon® processors with built-in Intel® Accelerator Engines. These are the first Azure instances on this generation of processors, bringing a significant leap in performance, cost reduction, and improved power efficiency for today's workloads.
The new public preview instances feature Intel® Advanced Matrix Extensions (Intel® AMX) – a built-in accelerator that significantly improves the performance of deep-learning training and inference on the CPU. This allows workloads like natural-language processing, recommendation systems, and image recognition to maintain high performance while running within a smaller trust boundary. To help you build and deploy models faster, Intel® AMX instructions are integrated into PyTorch and TensorFlow frameworks.
The new Dv6 VMs offer a balance of memory to vCPU ratio with increased scalability of up to 128vCPUs and 512 GB of RAM. Example workloads include many enterprise-grade applications, e-commerce systems, and entry-level and mid-range databases.
The new Ev6 VMs are ideal for memory-intensive workloads offering up to 192vCPU to 1832 GB of RAM. Example workloads include SAP, and more broadly memory-intensive enterprise applications, large relational database servers, data warehousing workloads, business intelligence applications, and in-memory analytics workloads.
The new instances come with three different memory-to-core ratios and offer options with and without local SSD across all the new VMs – the General Purpose Dsv6, Dlsv6, Ddvs6, and Dldsv6 series and the Memory Optimized Esv6 and Edsv6 series. Additionally, there are constrained core variants for Esv6 series which are ideal for workloads that require high data throughput without a high number of vCPUs.
These VMs are built on the 5th Gen Intel® Xeon® Platinum 8537C processor and offer:
- 20% higher vCPU performance and 3x larger L3 cache than the previous
generation Intel Dl/D/Ev5 VMs. - Up to 260k IOPS, 12.5GB/s remote storage throughput.
- Up to 200Gbps VM network bandwidth powered by Azure Boost.
- 46% larger local SSD capacity and >3X read IOPS.
- 84 new VM sizes in total, ranging from 2 to 192 vCPU and up to 1.8TB of memory.
- NVMe interface for local and remote disks for all sizes.
These new VMs significantly extend the performance of Intel VM options available for Azure customers.
All sizes with less than 1TB of memory are available for deployment in the US West and US East regions, with additional VMs (1TB or above of memory), coming into the preview in June or July 2024.
Azure customers can sign up for the public preview through the preview signup form. And to learn more, visit the Microsoft blog.
Notices and Disclaimers
Performance varies by use, configuration, and other factors. Learn more on the Performance Index site.
Performance results are based on testing as of dates shown in configurations and may not reflect all publicly available updates. See backup for configuration details. No product or component can be absolutely secure.
Your costs and results may vary.
Intel technologies may require enabled hardware, software, or service activation.
© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others.
Josh Hilliker is a Senior Cloud Solutions Architect in Data Center Sales focused on advancing our Autonomous Data Center solutions, including telemetry and infrastructure modernization. Josh has held many past roles at Intel including the following: Director in SMG University, Platform Marketing in Business Client Platform Division, Intel GE-Care Innovations as the Director of Social Media, Online Presence and Google Analytics for consumer products, Product Engineer with the IT Land Capital and Acquisitions team, Operations manager for Intel e-Business Infrastructure & OEM Business Link, Architect / Community manager for Intel® vPro ™ Expert Center, Intel Manufacturing, and Director of Small Business initiatives in Intel Architecture Group.