Showing results for 
Search instead for 
Did you mean: 

AMD EPYC™ Processors are ideal for HCI - from Edge to Core

4 0 6,995

In order to understand Hyperconverged Infrastructure (HCI), a brief history of the evolution of the legacy data center to the modern-day data center is needed. Legacy data centers are typically composed of a multi-tier architecture made up of a storage tier, a networking tier, and a compute tier. Each of these components would typically be managed by a different administrator using purpose-built hardware creating a natural barrier, or silo, because of the functionality and expertise required to manage them.


The traditional data center model has been in place for decades. Its rigidity and attendant inefficiencies led to a search for solutions culminating in the creation of Hyperconverged Infrastructure (HCI). Initially, HCI was thought of as a type of software-defined storage, primarily because software abstraction of the traditional enterprise storage architecture was the last element necessary for a truly software defined data center. HCI has grown to be much larger than its original scope, combining server virtualization with software defined networking and continuous availability through self-healing along with advanced management and analytics capabilities.


Today, HCI is mainstream and offers a cloud experience in the customers own data center, bringing efficiencies and agility for demanding IT requirements. AMD has been working closely with our ecosystem partners in delivering fully tested and validated solutions that offer outstanding performance, scalability and TCO. While this is not an exhaustive list, these are some of the most common HCI use cases:


  • General Purpose Computing – Virtualization had already started a trend towards server consolidation. But significant deployment planning was still required to avoid stress on existing storage and network infrastructure. By bringing storage within the node and distributing it across the cluster, much of this overhead could be avoided, and server consolidation can continue in an easy and predictable fashion. One of the unique differentiations that 2nd Gen AMD EPYC processors offers is the core density – up to 64 cores and 128 threads per processor – which enables the ability to run higher virtual machine density and while reducing infrastructure.
  • Virtualized Databases – Databases and other Tier 1 applications are finding that HCI can provide enough performance for these workloads. Historically, it was assumed that HCI would be unable to meet these needs, but with recent advances in HCI technology along with performance and feature enhancements that AMD has brought to the table, such as higher base and boost frequency processors, along with high speed I/O and network enabled through PCIe® 4, this is no longer the case. It’s also important to note that the memory capacity advantage that 2nd Gen AMD EPYC processors has – up to 4TB of memory per processor - can accelerate in-memory computing for transactional and real-time analytics workloads. That is not to say that all business-critical applications are a good fit for HCI yet, but even some of the traditional database vendors are starting to see the appeal.
  • Virtual Desktop Infrastructure (VDI) – VDI historically pushes virtualized servers to the very edge of their capabilities. Today, VDI is bringing an even richer user experience to a mobile and distributed workforce. VDI enhances centralized control and protection over business-critical data while supporting collaboration. A better user experience is tied directly to server capability. Industry-leading core count coupled with the high memory capacity and bandwidth in AMD EPYC processors enables optimal virtual desktop density and performance.
  • Edge Computing – HCI is becoming a popular choice for Remote Branch Office and Edge Computing. Traditional systems are overkill, being both too costly and too complex for such deployments. AMD EPYC processors offers the power and space efficiencies required for edge environments. With AMD EPYC processors, these self-contained small data centers can be deployed efficiently at a fraction of the cost. Beyond deployment advantages, the core capabilities of HCI, such as provisioning, monitoring, management and on-demand scaling, can significantly reduce the complexities associated with edge computing.


Nutanix is a world leader in HCI and we at AMD are very excited to collaborate with them. We have worked together on optimizing the Nutanix hyperconverged software, Acropolis OS, on AMD EPYC processors. Together, we have enabled choice in hypervisors including Nutanix Acropolis Hypervisor (AHV), VMware ESXi®, and Microsoft® Hyper-V. We have been collaborating closely with our server OEM partners in bringing fully validated HCI solutions including the Nutanix-integrated HPE ProLiant DX385 appliance, the Dell EMC XC Core XC-6515 and newly announced Lenovo ThinkAgile HX HCI solution.


We are a proud sponsor of Nutanix Global.NEXT 2020 and look forward to helping you meet today’s business challenges with fully validated hyperconverged infrastructure solutions. You can learn more about AMD EPYC processors for Nutanix solutions at the Partner Xchange Breakout Sessions and here at the AMD website.


Raghu Nambiar is a CVP of Datacenter Ecosystems & Application Engineering for AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites or use of third party names/marks are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

About the Author
Raghu Nambiar is the Corporate Vice President of Datacenter Ecosystems and Solutions at AMD. In this role, he leads engineering teams and their collaboration with ecosystem partners. Raghu has more than 20 years of technology industry experience across a number of engineering organizations. He was previously the CTO of the Cisco UCS business and played an instrumental role in accelerating the growth of the Cisco UCS to a top data center compute platform. He has spent his entire career working on software and hardware ecosystems for data centers, both on in research and business use cases.