Announcing Leading-edge Architectural Innovation for VMware Cloud on AWS to Support Next-gen Workloads

Today, at AWS re:Invent 2023, we announced a groundbreaking disaggregated infrastructure option for VMware Cloud on AWS. Soon, customers will be able to deploy a brand new m7i.metal-24xl instance and have the flexibility to connect either VMware Cloud Flex Storage or Amazon FSx for NetApp ONTAP external NFS storage within their Software-Defined Data Center (SDDC).

With this significant architectural innovation, we break the link between compute and storage. This innovation empowers customers to choose their storage and compute options independently, optimizing infrastructure costs for specific use cases. With support for the disaggregated M7i and the hyperconverged i4i and i3en instances, we offer unprecedented flexibility by providing customers with a spectrum of choices for allocating and scaling their compute and storage capacity. Furthermore, with the introduction of M7i instance, customers will be able to accelerate their AI initiatives by deploying modern AI/ML workloads on VMware Cloud on AWS.

The introduction of the bare metal M7i instance is just the initial stride in taking advantage of the new disaggregated architecture, setting the stage for a diverse range of new instance types in the future.

So, let’s take a closer look into this new instance type- m7i.metal-24xl:

Why do customers need this?

In the ever-changing landscape of modern business, IT teams face the crucial task of managing infrastructure capacity with flexibility and efficiency. Therefore, there’s a growing demand for more granular control over compute and storage resources. One way to address this need is by disaggregating compute from storage, which empowers IT teams to independently allocate and then scale both of these resources. The desire for flexibility extends to the need for a diverse array of storage and compute options to address various business needs.

Also, customers are seeking cost-effective infrastructure solutions, particularly for specific use cases where the balance between storage and compute resource consumption is paramount. The aim is to avoid unnecessary expenses by eliminating the burden of paying for unused or redundant capacity.

Furthermore, in today’s age of Artificial Intelligence, customers want to harness the power of Gen AI and for that, they need a secure, robust, high-performance infrastructure.

Solution: m7i.metal-24xl instance type

• An Amazon EC2 M7i.metal-24xl instance is powered by custom 4th Generation Intel Xeon Scalable processors (code named Sapphire Rapids) with an all-core turbo frequency up to 3.9 GHz.
• 4th Gen Intel Xeon Scalable processors feature Intel® Accelerator Engines designed to accelerate performance across the fastest-growing workloads.
• With Advanced Matrix Extensions accelerator, it accelerates matrix multiplication operations for deep learning (DL) inference and training workloads*.
• It offers:
  • 48 physical cores, 96 logical cores with Hyper Threading enabled.384 GiB memoryFlexible NFS storage options to choose from as per customer needs – VMware Cloud Flex Storage or Amazon FSx for NetApp ONTAP.
  • Up to 37.5 Gbps networking speed
• It supports always-on memory encryption using Intel Total Memory Encryption (TME)

Customer Benefits:

m7i.metal-24xl instance type’s disaggregated architecture provides more flexibility to customers in terms of compute/storage options. Customers can choose from one of the three instance types (i3en.metal, i4i.metal or m7i.metal-24xl) and different storage options such as vSAN, VMware Cloud Flex Storage or Amazon FSx for NetApp ONTAP to better align with their workload requirements.

Customers can scale the compute and storage independently (E.g., Scale m7i.metal-24xl with VMware Cloud Flex Storage or Amazon FSx for NetApp ONTAP) as requirements evolve in future.

For the use cases where customers need only limited amount of compute, storage and memory resources (e.g. Ransomware and Disaster Recovery), m7i.metal-24xl instance type provides a cost-effective infrastructure with smaller resource configuration for these types of use cases.  Also, customers can granularly control storage consumption and optimize the storage costs based on their needs. (e.g. with vSAN customers are required to add a host with 20TiB storage while with Amazon FSx for NetApp ONTAP, customers can add storage as small as 100GiB)

The 4th Gen Intel Xeon Scalable processor offers the CPU with the most built-in accelerators that helps improve performance efficiency for emerging workloads such as AI/ML workloads, analytics or HPC workloads. It offers up to 1.53x average performance gain over the previous generation processors¹

Use Cases:

Some of the key use cases of m7i.metal-24xl instance are:

• CPU Intensive workloads
• AI/ML Workloads
  • Natural Language Processing
  • Classical Machine Learning
• Workloads with limited resource requirements
  • Web and application servers
  • Microservices/Management services
  • Secondary data stores/database applications
• Ransomware & Disaster Recovery
  • Modern Ransomware Recovery
  • Compliance and Risk Management

What’s Next?

If you would like to take a sneak peek into how it looks like, please check out a cool demo we have developed. Also, please check out technical deep-dive information about m7i.metal-24xl in this TechZone article- A Closer Look at the Diskless Instance on VMware Cloud on AWS.

For any further questions, please reach out to your VMware or AWS representative.

If you are at AWS re:Invent, don’t forget to attend the sessions on VMware Cloud on AWS mentioned in this blog where you will learn more about this announcement. Also you can schedule a meeting with VMware SMEs by submitting the request here- Request a meeting

Resources:

If you would like to learn more about VMware Cloud on AWS or any of the NFS datastore options supported on M7i, please check out the resources below:

* Requires virtual Hardware version 20

¹ https://www.intel.com/content/www/us/en/products/docs/processors/xeon-accelerated/4th-gen-xeon-scalable-processors-product-brief.html