Arista Scales Up Ethernet Portfolio with Big Switch

By Michael Feldman

April 20, 2010

The axis of datacenter networks is shifting and for companies like Arista Networks and other high performance 10 Gigabit Ethernet (GbE) switch vendors, this is opportunity knocking. The North-South network traffic pattern of client-server computing that dominated the industry for so long is giving way to high performance East-West networks where server-to-server communication is paramount. And its the rise of the HPC and cloud computing markets that are helping to drive this realignment.

On Monday at the High Performance Computing Linux Financial Markets conference in New York, Ethernet switch vendor Arista Networks announced its first modular switch for the datacenter. The Arista 7500 provides 384 wirespeed L2/L3 10GbE ports in a compact 11RU chassis that the company is touting as “the highest throughput 10 Gigabit Ethernet switch in the industry.” With an aggregate switching capacity of 10 Tbps, a packet memory of 18 GB, a port-to-port latency of 4 to 10 microseconds, and the ability to sling 5.7 billion packets per second, the 7500 is certainly a switch to be reckoned with.

According to Anshul Sadana, Arista’s VP of Systems Engineering and a Cisco alum who joined the company in 2007, they’ve leapfrogged their Ethernet competition. “Even their roadmap slide decks haven’t caught up with what we’re delivering to the market,” he says.  In fact, Arista is offering 40 and 100 GbE upgradeability on today’s 7500 platform. So when the rest of the ecosystem catches up, a fully tricked out 7500 switch could be transformed into a 48-port 100GbE box

Until the 7500 was launched this week, the company portfolio was restricted to its 24- and 48-port rack switches — the 7048, the 7100 T and 7100 S — that sit at the leaf layer of the network. The new 7500 is aimed at the spine layer – the super-highway that aggregates the network traffic from the compute and storage leaf switches. Like all of Arista’s gear, the 7500 is designed for datacenter applications that require a low-latency, non-blocking network fabric. In the past, this type of setup was confined mostly to HPC, but today, with the build-out of cloud computing infrastructure, it’s on its way to becoming a much more widespread networking model.

Arista is one those companies that is riding the GbE to 10GbE industry transition. After less than three years from its launch, Arista has already managed to accumulate over 300 customers spread across more than 25 countries. Sadana says the company is not revealing revenue figures at this point, but claims they are “growing exponentially.”

Currently, about 30 percent of their customers are financial institutions within the high frequency trading space, where Arista’s ultra-low-latency 7100 S switches are a nice fit for applications that need to execute split-second trades. Although none of these customers are admitting to using Arista gear, Sadana says you’ll see plenty of their switches in just about any co-location facility next to the major stock exchanges. According to him, some of these deployments are in the 100-switch range, deployed across multiple sites. “I would say we’re the number one vendor in that space today,” says Sadana.

Arista’s other customers are spread across the oil and gas industry, universities, US DOE labs, Web 2.0 datacenters, and cloud/hosting providers.  Specific customers — at least ones who will publicly admit to owing Arista gear — include the National Center for Atmospheric Research (NCAR), SARA in Europe, the University of Colorado, the BBC, Lawrence Livermore National Laboratories, and the San Diego Supercomputing Center, among others

Some of the first customers of the 7500 switch include the Howard Hughes Medical Institute, the San Diego Super Computer Center (SDSC) and a yet-to-be named DOE lab. Early testing on a 504-node (4,032-core) cluster employing an Arista 7508 switch and 7148SX top-of-rack switches yielded a Linpack performance mark of 35.8 teraflops. That represents 84 percent of the machine’s peak performance, which is on par with the Linpack efficiency of InfiniBand-connected clusters. Only proprietary interconnects, such the one in Japan’s Earth Simulator supercomputer can do much better — 93 percent efficiency, in this case.

In fact, Arista switching pretty much matches InfiniBand performance and utility in most areas except perhaps raw bandwidth. But compared to traditional Ethernet aggregation switches, the 7500 boasts much better performance, energy efficiency, and price-performance. For example, Cisco’s Nexus 7010 switch can only support 64 10GbE ports at wirespeed and just under 1 billion packets per second. And at 21RU, the Cisco switch is nearly twice as large as the Arista chassis.

A fully configured 7500 is listed at $460,000, which works out to around $1200 per port. That’s 1/30 the cost of its Cisco switch counterpart. Furthermore, since the Nexus box consumes 139 watts per wirespeed 10GbE port, compared to 13.2 watts for the Arista gear, energy costs for the 7500 are going to be about 1/10 that of the 7010.

To be fair, the Nexus 7010 incorporates a lot more routing capability and other features that makes it a more capable end-to-end switch for the datacenter. But that was part of Arista’s calculation. By focusing on the spine layer of the network, the company could maximize aggregate bandwidth and minimize latency, while beating the competition significantly in cost and power consumption.

In the short term, more formidable competition may come from InfiniBand vendors — and not just from InfiniBand switches. Last year, Voltaire launched the Vantage 8500, an Ethernet core switch with a capacity of 11.5 Tbps that provides up to 288 10GbE wirespeed ports in a 15U chassis. The 8500 is an L2 core switch, so presumably it lacks some of the L3 routing smarts of the Arista 7500. But since Voltaire employs cut-through rather than the store-and-forward switching of the Arista 7500, the 8500 is able to keep latencies in the 1 microsecond range. (On the other hand, store-and-forward has some performance advantages in the spine layer, especially for storage clusters and TCP traffic.) Power consumption and pricing per port for the 8500 are comparable to the Arista solution.

On Monday, BLADE Network Technologies and Voltaire announced they are combining forces to offer a lossless converged Ethernet fabric solution based on the 8500 and BLADE’s G8124 24-port top-of-rack switch. The BLADE-Voltaire solution is aimed at applications that rely on the same type of low latency, non-blocking networks that Arista is targeting, namely HPC and cloud-scale apps. The switch combo is packaged with Voltaire’s Unified Fabric Manager to manage switch resources, and is designed to scale up to 3,400 10GbE ports. The solution also comes with BLADE’s VMready (for virtualized environments) and Voltaire Messaging Accelerator software (for high frequency trading applications).

Although InfiniBand is still the interconnect of choice where the lowest latency and highest bandwidth are needed, solutions like those from BLADE, Voltaire, and Arista will make it easier for HPC, high frequency trading, and ultra-scale datacenter applications to transition from GbE to 10GbE networks. And with 40 GbE and 100 GbE just over the horizon, the protocol will be with us for the foreseeable future.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

Nvidia Appoints Andy Grant as EMEA Director of Supercomputing, Higher Education, and AI

March 22, 2024

Nvidia recently appointed Andy Grant as Director, Supercomputing, Higher Education, and AI for Europe, the Middle East, and Africa (EMEA). With over 25 years of high-performance computing (HPC) experience, Grant brings a Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Houston We Have a Solution: Addressing the HPC and Tech Talent Gap

March 15, 2024

Generations of Houstonian teachers, counselors, and parents have either worked in the aerospace industry or know people who do - the prospect of entering the fi Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire