InfiniBand Rides HPC Wave Into the Enterprise

By Michael Feldman

June 1, 2007

Last week, IDC released a report that projects a rather healthy future for InfiniBand (IB) adoption over the next five years. The study predicts that InfiniBand host channel adaptor factory revenues will grow from $62.3 million in 2006 to $224.7 million in 2011, while InfiniBand switch port sales are expected to grow from $94.9 million to $612.2 million over the same period. By the end of the decade, the deployments of double data rate and quad data rate InfiniBand adapters are expected to overtake their single data rate forbearers.

Compared to Ethernet or Fibre Channel, InfiniBand has the advantage of providing a high bandwidth, low latency interconnect in a less expensive package. Current implementations are being delivered at 10, 20 and 40 Gbps, with products in the works for 120 Gbps. Its superior performance has made it the interconnect of choice in many HPC deployments. According to the latest Top500 list, 12 percent of the top systems use InfiniBand, up from 5 percent just a year ago. By itself , this statistic is largely irrelevant. The fact that 60 or so of the fastest computers use a particular interconnect may give the IB vendors some bragging rights, but it doesn't say much about overall industry adoption. What is important is that InfiniBand's use is expanding in commercial cluster computing, where interconnect performance and price-performance is the driving factor. This includes such applications as automotive crash simulations, oil and gas reservoir simulations, and financial analytics. These kinds of computing workloads have become mainstream as HPC has become a cost-effective tool for a wide array of businesses.

But much of the projected growth of InfiniBand is expected outside of its traditional role in connecting high performance computing servers and storage — in mainstream datacenters. The IDC report makes the case that the growth of multicore processors, server virtualization, and I/O consolidation is going to drive InfiniBand adoption in the enterprise. All three of these trends are helping to make systems more computationally dense, which requires proportionately more communication bandwidth per server and per storage device. Besides raw performance, InfiniBand includes quality-of-service (QoS) features that enable multiple types of traffic to be safely managed over a single pipe.

Over the past few years, more mainstream enterprise users have started using InfiniBand. This is especially apparent in the capital markets, where the goal for ever-faster automated trading is on a collision course with increasing trade volumes. In this environment, even sub-second delays in transactions can cost millions of dollars. The types of computing systems that manage these trades have come under increased scrutiny as financial engineers ponder how best to minimize transaction latencies and enhance trade predictibility. Wombat Financial Software and Reuters are two companies that have qualified their market feed applications on InfiniBand technology to address these stringent performance requirements. InfiniBand's market penetration in this sector is largely unknown, since financial institutions tend to be rather tight-lipped about what goes on inside their datacenters. But with so much money in the balance, one can assume that all market trading institutions are taking a hard look at InfiniBand.

SOA platforms can benefit from high performance interconnects too. In May, TIBCO Software announced it had qualified its message passing middleware on top of Cisco's InfiniBand Server Fabric Switches to enhance performance and predictability for event-driven SOA. The types of applications targeted include data distribution, web services and order management systems. Order management, in particular, is becoming a time-critical component of inventory control for many companies. TIBCO claims their IB-enabled platform increased throughput by a factor of four, while reducing latency.

Database clustering, as is used in Oracle's 10G RAC and IBM's DB2, is another prime target. Following the general model of concentrating computational power into smaller boxes, database clusters are becoming more dependent on high performance interconnects to talk between the nodes. For databases applications, there is the additional incentive to unify the interconnect fabric with the storage components. JDA, a provider of supply chain management software, is using InfiniBand to improve system throughput on its platform. Using an Oracle 10G RAC system and QLogic InfiniBand gear, the company was able to decrease the time to plan one million SKUs (Stock Keep Units) from 66 minutes, with Gigabit Ethernet, to 25 minutes, with InfiniBand. They were also able to realize a 35 percent cost advantage by switching from Ethernet to InfiniBand.

It's no surprise that as the application and database tiers in the datacenter start to act like HPC systems, they will start to look like them. In doing so, we're bound to see InfiniBand adoption increase in the larger enterprise market. Since IB technology is projected to maintain its performance and price-performance advantage for at least the next five years, users will look to InfiniBand for their most challenging interconnect demands.

That's not to say Ethernet or Fibre Channel are going away. Both are established standards and have momentum that will carry them for a long time. The massive set of applications that are run on standard Ethernet will insure its longevity. For the foreseeable future, Ethernet is expected to represent the most common networking technology in the world. The web tier of the enterprise is one area where Ethernet has no serious challengers. And if you've been reading this publication for any length of time, you know that creative engineers have been devising new ways to enhance the performance and predictibility of Ethernet.

The culture of the InfiniBand vendors has matured as well. There's little talk of InfiniBand conquering the world these days. Even IB evangelists like Mellanox and Voltaire are shipping Ethernet products. Over the past couple of years, acquisition of companies like Topspin (Cisco), PathScale and Silverstorm (QLogic) has brought InfiniBand into more enterprise-focused vendors. The realization that all the standard network interconnects will live peaceably in the datacenter may be sinking in.

“From the Mellanox perspective we have technology that will address both markets,” says Thad Omura, Mellanox VP of Product Marketing. “We're layering one growth market on top of another. It's no secret that 10 Gigabit Ethernet will emerge at some point. There's no reason why we shouldn't address the 10Gig solution for LANs and continue to drive InfiniBand as the best price-performance interconnect for servers and storage.”

—–

As always, comments about HPCwire are welcomed and encouraged. Write to me, Michael Feldman, at [email protected].

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!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues 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 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…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion 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…

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