iGrid Chair DeFanti: ‘Not at the End of the Rainbow Yet’

By By Doug Ramsey, University of California, San Diego

September 12, 2005

In a recent interview for GRIDtoday, Doug Ramsey of the University of California, San Diego spoke with Tom DeFanti about the upcoming iGrid 2005 workshop and symposium, which will be held Sept. 26-30 in San Diego.

DeFanti — iGrid 2005 co-chair, co-director of the Electronic Visualization Laboratory at the University of Illinois-Chicago, and senior research scientist with the UCSD division of Calit2 — began by joking: “Sometimes, I think of it as the visualization, networking and Grid computing equivalent of a Grateful Dead concert.”

GRIDtoday: How so?

TOM DEFANTI: This is not a typical conference. It's really a workshop, and almost everyone who attends is contributing in one way or another, and more than half are contributing content. We see iGrid as a very intense workshop to develop new ideas and capabilities, and people risk a lot in terms of trying things that have never been done before. Many of the demonstrations will also be shown at Supercomputing 2005 in November, and when they arrive at SC, they want to be sure everything works. So at iGrid, it is okay if something doesn't work, because we know we have six weeks more to make it work. Part of being a scientist is that the answers are not anywhere near as interesting as the questions. A workshop like iGrid provides enough answers to formulate the next questions.

Gt: Can you trace the geneology of iGrid for us?

DEFANTI: The whole idea started when we networked McCormick Place in Chicago and did remote visualization as part of SIGGRAPH 1992. Three years later we connected IWAY at Supercomputing in San Diego, and it was important because it was the first time anyone had gotten the federal networks to talk to each other. That was just 10 years ago. It was also one of the first conferences where the Web was used extensively. We also learned how to build this international team, so we staged the first iGrid at Supercomputing 1998 in Orlando, Fla. Because we wanted to emphasize international collaboration, we decided to push our luck and have the 2000 iGrid in Yokohama, Japan, then in Amsterdam in 2002.

Gt: The vast majority of applications demonstrated at iGrid 2005 will also be international. Why that emphasis?

DEFANTI: The Supercomputing conference has always been fundamentally a U.S.-focused show. But for iGrid, almost all of the demonstrations have some international component because we were and are funded by the National Science Foundation to support science and engineering education across international networks. The idea was to give people in the United States and their international colleagues a chance to show off their stuff in a supportive environment. Obviously, a real huge issue is to keep the U.S. preeminent in a sector where funding is rising rapidly in other countries such as Canada, Korea, Japan and China. The European Union's funding as a group way outstrips U.S. funding in this area. We used to be the only act in town, but now we're part of a global community and struggling to keep up.

Gt: How cohesive is the international networking community?

DEFANTI: To our great surprise and delight, when we did the program in Yokohama it went very smoothly. Although there are many cultural and other things that are very different in Japan and Asia, the networking world is the same. They speak the same language. The group of people we deal with — technologists, scientists and engineers — all work together and like it. We discovered that communicating with our Pacific Rim counterparts was no problem. They have the same technology and the same culture of technology. The same holds true in Europe.

Gt: What is the core networking technology driving iGrid 2005?

DEFANTI: In 2002 in Amsterdam, we had 10-Gigabit circuits, so it was the first time we had more bandwidth than we needed. But now we can control lambdas, these wavelengths of light, so it's about user control and application control of the lightpaths. Several different demonstrations will involve controlling lightpaths and bringing things in simultaneously from all over the world. There are also demonstrations of equipment that does that, and applications that use the extreme bandwidth. This bandwidth essentially now exceeds the capacity of computers talking to each other singly, so you wind up having parallel communications going on. Many of the experiments rely on the traditional Internet, but a surprising number use lightpaths.

Gt: Whose networking equipment will be used during iGrid?

DEFANTI: We received a very nice donation from Cisco Systems, and we have loaner equipment from Force10. The Cisco Catalyst switch and the Force10 switch are the heart of our standard networking, and we also got loaners from Nortel Networks to do the Layer 1 lightpath work, which in the past year has become the basis of the next phase of international networking. We are hoping that now we have this capability here, the campus and the UC system will commit to making San Diego one of the core hubs for lightpath networks, along with Chicago, New York, Amsterdam, London, Tokyo and other Asian open Internet exchanges.

San Diego is a rational place to install this capability on a permanent basis, with both Calit2 and the San Diego Supercomputer Center here. For iGrid, the networking in San Diego will connect through the national CAVEwave, PacificWave and National LambdaRail to international connections to the east and west, and through CENIC to UC-Irvine, University of Southern California and other California destinations. These aren't networks that people depend on for their banking, so we can try adventurous things and floor the accelerator without worrying about inconveniencing anybody.

Gt: Although iGrid will use the latest networking technology, most of the demonstrations will be application-based, right?

DEFANTI: It is fundamentally focused on applications and content, and content is king. People want to spend money on applications, they don't spend money on computer hardware for hardware's sake. Scientists will use our technology if it helps them do better science. So iGrid is very focused on applications, to show what people are doing in a world of almost limitless bandwidth.

Gt: What are some of the recurring themes among applications to be demonstrated at iGrid?

DEFANTI: One that has emerged this year is high-definition television (HDTV) over lightpaths, including streaming HDTV and interactive HDTV. Attendees will also witness a new industry standard called 4K. At 4096 by 2180 pixels, 4K is over four times the definition of regular HDTV. Sony is lending us one of their first 4K projectors, and NTT will stream 4K content from Japan to San Diego. And professor Peter Otto is working with Skywalker Sound to synchronize audio for these 4K demonstrations, and that should be very impressive. More and more people are using this technology to see and interact with data coming from sensors, and an HD camera can be a sensor. With the drop in price of HD cameras from $100,000 to as little as $3,000, and with networking increasingly built into these cameras, you can just plug these things into a network and off you go. This will be a tremendous push on network capacity, because an HD camera exceeds a Gigabit if you don't compress it, and although 4K imagery can be compressed down to under a Gigabit, it's huge in uncompressed form, and you want to keep stuff uncompressed if you want to edit it or do post-production effects on it.

Gt: How important is security as an issue for this year's applications?

DEFANTI: There is a lot of attention on security. Whether they are dealing with medical images, or Hollywood movies, or remote visualization for the military, all of these groups have extreme needs for security for different reasons. One is legislated, one has to do with intellectual property and money, and the other is for national security. They are all very important, and increasingly people want to move security to the network and away from the end user. I devote way too much time every week to keeping my laptop safe, and I shouldn't have to. We are trying to solve that problem for high-end images, so we are doing an advanced security demo with Canada's Nortel Labs, where they build the security into their switches. You wouldn't even know it's there, but if someone tapped into the line, it looks like gibberish.

Gt: So iGrid goes well beyond what many people would label 'Grid' computing or networking?

DEFANTI: The Grid should not only control computing and disk-space resources, but also networking resources. To be clear, we call it the LambdaGrid because it goes beyond the traditional Grid that uses the shared Internet, into the realm of dedicated lightpaths, or lambdas. The LambdaGrid gives scientists the option of having schedulable or dedicated resources that are shared in some way, and the cost of these resources is no longer the stumbling block. These networks are expensive, but not extremely expensive any more. If you take all the networks that we're putting into iGrid and you pay for them, it's a fraction of the cost of the people who are using them. A national or international network doesn't cost much more than a full professor or a networking technician per year, so that changes the equation.

Gt: What draws so many of your colleagues from around the world to iGrid?

DEFANTI: Over the years, we have developed an international community of people who trust each other and who see the benefit of teamwork with those in visualization, in networking, in computing, in simulation, or in HD video. We lend and send each other pieces of equipment that cost $50,000 or $500,000, and we know we'll get them back. We trust each other with our technology and with some of the secrets that we hold dearest. This is kind of our celebration of how much we trust each other. Maxine Brown and I have been running events for a long time, and like we did with SIGGRAPH, we want iGrid to be a place where groups of people could show their exquisite technology to fellow scientists. In iGrid, we are trying to get people to bring stuff out of the lab and share it with each other. It's not like a regular conference where you get up and show PowerPoints. There are almost no canned presentations, although there will be a symposium running concurrently to round out the information. But it's not the dominant reason people come. They come to see the technology and applications in action.

Gt: Since visualization is very important at iGrid, what other technologies will be used to showcase individual applications?

DEFANTI: Apart from Sony's 4K projector in the Calit2 digital cinema, we have multiple, high-intensity projectors that are 1600×1200 pixels. We have a big stereo screen in a smaller auditorium, and we're building a 100-megapixel screen in another. We will also have a desktop virtual-reality device that shows 3-D images but requires no special glasses and tracks your position by neural networks and video. Important to the HDTV theme, NASA is bringing a personal HDTV stereo device that also requires no glasses. We also have two setups with lovely 63-inch Samsung plasma panels. These are all facilities so people can show their work in the best way possible, but they also serve another purpose: 49 different groups will be showcasing their applications on a total of 10 different ‘screens,' so they will be sharing the visualization facilities. Sharing is not an inconsiderable obstacle when you are pushing the limit of the technology, so these groups will hopefully learn from each other.

Gt: You are a co-principal investigator on the Calit2-led OptIPuter project. What do iGrid, Calit2 and the OptIPuter project have in common?

DEFANTI:
We decided to hold iGrid in the new Calit2 building to force the immense capability and infrastructure of the building into functional being in a huge rush. By rights, it should have taken a year to bring this sort of network infrastructure into a 215,000-square-foot building like this, but we're doing it in weeks. As for OptIPuter and iGrid, they have similar goals. OptIPuter is using LambdaGrids to help scientists see and move their data. But iGrid is a superset of the OptIPuter: there are other people at iGrid with ideas — some complementary, some competing — on how to architect this world where bandwidth is no longer a constraint. We hope to witness applications and technologies at iGrid that can be used in the OptIPuter. We are all learning from one another, and as these things evolve into real software products, companies will adopt the best of each technology. We're not at the end of the rainbow yet.

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