Hi. Welcome to Ingram Micro.

Please choose your role, so we can direct you to what you’re looking for.

If you’d like to learn more about Ingram Micro global initiatives and operations, visit ingrammicro.com.

Ethernet Switching for Enterprise Data Centers

November 12, 2021

Ethernet Switching for Enterprise Data Centers
powered by Sounder

When it comes to data center expansion and cloud migration, NVIDIA Spectrum Switches might be exactly what you need.

Shelby Skrhak speaks with Jeff Shao, Director of Ethernet Alliances, and Bill Webb, Director of Ethernet Switching - Americas, both with NVIDIA, about:

  • The three pillars of data center expansion
  • Why NVIDIA Ethernet Storage Fabric is important to modern data centers
  • Joint solutions through NVIDIA’s ecosystem of partners

Three pillars

“The data center is now the new unit of compute,” Bill says, “and because of that, the network is really critical to the modern data center.”

NVIDIA Spectrum Switches are tailor-made for this new data center. In fact, they support the three pillars of data center expansion.

1. Operational efficiency

The switches use Cumulus Linux, which is great for automation.

Plus, they offer a digital twins environment that allows customers to develop their automation, fully validate it before rollout and use the same automation tools to roll it out.

These aspects accelerate deployment of networks and prevent problems that oftentimes delay expansion.

2. Hardware efficiency

It’s critical to keep up with the hardware demands of the modern data center. NVIDIA has been building high-performance networks for over 20 years and are well versed in helping you get the most out of your investments.

3. Visibility/telemetry

NVIDIA helps speed up mean time to resolution. If something goes wrong, NVIDIA’s technology will send instant root-cause information.

Ethernet Storage Fabric

“In today's data centers, big monolithic storage systems are being replaced by disaggregated, scale-out network storage, where storage in a cluster of servers is put to service applications in a much more agile, scalable and cost effective way,” Jeff says.

Because this data is now distributed, there’s a lot of back and forth between storage and compute servers.

“To improve efficiency, we converge compute and storage in the same rack or even on the same server,” Jeff says.

The result is converged and hyperconverged infrastructure.

Performance is very critical in these new infrastructures, so a single unified storage network that can support all data types and applications, and maximizes the capability of storage and compute service is needed.

Plus, the storage network needs to be highly scalable, reliable and automated for data center expansion.

That describes Ethernet networking to a tee.

The Ethernet storage network that uses a leaf-spine architecture is also known as Ethernet Storage Fabric or ESF.

In this architecture, there are two layers of switches:

  • A layer of leaf switches in the racks providing server connectivity
  • The upper layer of spine switches connecting those racks to the external network

“Just remember that ESF is the data center storage network that is simple, high-performing and easy to scale,” Jeff says.

Joint solutions

NVIDIA has built a very broad ecosystem of partners with leading data center and cloud solution providers.

“By leveraging these unique capabilities, our partners and media are delivering integrated and cost effective solutions to our data center customers,” Jeff says.

To join the discussion, follow us on Twitter @IngramTechSol #B2BTechTalk

 Listen to this episode and more like it by subscribing to B2B Tech Talk on Spotify, Apple Podcasts, or Stitcher. Or, tune in on our website.