SCinet Satisfies the Need for Speed

By Michael Feldman

July 7, 2006

Summer has barely begun and some in the HPC community are already looking toward November’s 2006 Supercomputing Conference (SC06), which takes place in Tampa, Florida, this year. In fact, the effort to develop the conference network, SCinet, actually began at the end of 2005 and will continue throughout 2006, right up to the beginning of the conference.

SCinet supports the conference’s entire network requirements. And as one might imagine from a supercomputing event, it is not just a vanilla LAN. It has to provide a typical commodity network for the use of attendees, as well as a high-speed, state-of-the-art communication test-bed for various HPC research exhibits, demonstrations and event programs.

Over 100 volunteers from a large partnership of academic institutions, national labs, supercomputing centers, network hardware vendors, and telecommunications carriers are working together to design and build SCinet. The hardware vendors and carriers donate much of the equipment and services needed to build the infrastructure. While the planning typically begins more than a year in advance of each conference, the actual installation is done in the week just prior to the event.

“In that week, basically everything is built,” says Dennis Duke, chairman of SCinet for 2006. “All of the equipment is delivered, 60 to 70 miles of fiber are run, and all of the wide area network connections are made.”

Although most of the installation is done just before the event, Duke admits that at last year’s conference, the wide area network setup was actually started a month in advance because there were so many WAN connections. And it looks like they’re going to be doing the same thing this year. But the vast majority of the SCinet is installed in just seven days. Duke says the only reason this is possible is that they have several dozen of the best network people in the world there to build it — along with the support of all the vendors.

The vendors give both their time and their equipment to SCinet. According to Duke, the value of the donated equipment for each of last two years was around 25 million dollars. They know that because all the hardware has to be insured.

SCinet is actually composed of three networks:

1. The commodity network. This is the conference’s production network that is intended to be extremely stable and reliable and is similar to a network found at any research institution. It includes free wireless access for all the attendees and Gigabit Ethernet drops to all the booths and meeting rooms.

2. The high-performance network. This network is used to support high-performance demonstrations, the HPC Bandwidth Challenge, and other research exhibits. It will deliver multiple 10 Gigabit Ethernet links.

3. Xnet. This represents the conference’s bleeding-edge network. It will showcase experimental next-generation technology from vendors with equipment that is not quite ready for prime time. Unlike the commodity and high-performance networks, this infrastructure is not intended to be stable.

The Xnet is always the wild card in SCinet. No one knows what it’s going to look like until fairly close to the conference. Parts of it may be connected to the commodity and high-performance networks.

“The rule is that it can’t do anything that would endanger the stability of the other networks,” says Duke. “Apart from that, they can do anything they want. Last year, they set up an InfiniBand network that actually carried some wide area network traffic — very dramatic and successful.”

The HPC Bandwidth Challenge always seems to attract a lot of attention. This year they plan to do something a little different. Rather than focusing on pure speed, the emphasis in Tampa will be on production-level networking.

“So we’re not trying to create a record for how much bandwidth is used as much we’re trying to create a record for how much real-world work gets done, that is, production network capability,” explains Duke. “It’s sort of like the difference between peak performance and sustained performance on real applications.”

Duke says they are planning to have ten to twelve 10-Gig links into the Tampa Convention Center. In the past, each team typically had dedicated 10-Gig lambdas just for the Bandwidth Challenge. This year, each team will have a single 10 Gigabit link. So the teams are being encouraged to use their production networks at their own institutions, which means that in many cases upgrades will be needed. But it also means that once they are finished, they will have that high-performance backbone in place. Duke is hoping that the Bandwidth Challenge evolves into something that results in upgrading network capability all over the country.

The commodity network will also take on a new dimension this year. A high-performance wireless link will be added to accommodate the spill-over of the SC06 programs to buildings outside the Convention Center. For example, the Education Program will be hosted at the nearby Marriot Hotel, which is located about 200 yards from the Convention Center. SCinet will be linked to the hotel using a 2.6 Gigabits-per-second wireless beam from a company called GigaBeam. This state-of-the-art wireless technology represents a major speed bump for SCinet. By comparison, last year’s wireless link was a paltry 50 Megabits-per-second.

“That’s a first for us and the conference,” says Duke.

One of the biggest problems every year is upgrading the local infrastructure to be able to accommodate all the bandwidth required by the conference. Most cities do not have the type of network connectivity that is needed for something like SCinet.

“There’s a geographical challenge every year,” says Duke. “Wherever we go, they never have the local infrastructure to support what we need, because we’re effectively building one of the most powerful networks on the face of the earth. A good bit of effort during the year is to repair that problem.”

This year in Tampa it will be no different. The wide area network connection for the city comes into two places, about 12 miles from the downtown area. Both Level 3 and Quest will be spending a lot of their own resources to get the fiber downtown to within six blocks of the Tampa Convention Center. Verizon has also joined the SCinet effort this year as a major partner. They have installed a bunch of fiber pairs from the downtown POP into the Convention Center. According to Duke, SCinet precipitates this kind of major uprade every year.

Even within the convention centers, there are challenges. Most centers are geared for “normal” conferences and the local network is rather limited in performance. So for the past several, they’ve also had to rebuild the fiber infrastructures within the centers themselves. As SC has traveled around the country — Dallas, Denver, Baltimore, Phoenix, Pittsburgh and Seattle — they have left behind some very well-connected facilities. But the individuals who do all this work get something as well.

“The people who do it just love it,” says Duke. “It’s just such a challenge for them. That’s why they volunteer.”

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