|
|
||
|
23 June 2001 Live issue from the Heidelberg Supercomputer Conference, June 2001Primeur Live! is published during majorSupercomputing and Grid events in Europe
advertisement:
The Heidelberg Supercomputer Seminar is the main HPCN event in Europe. This year we publish two live issues from the event:
advertisement:
|
Where is supercomputing heading? Well, to know that, you would go to the Mannheim Supercomputer conference in June and be completely up to date. This year the conference moved to Heidelberg, and the now completely international conference attracted more participants than ever before: over 380. So a lot of supercomputer experts gathered there. But not a clear picture, not a direction. There must be something in the air, but where is it what is it? Several presentations pointed at the limits that could prevent further developments in the supercomputer field. Physical limits for instance, that prevent miniaturisation to go on. With shrinking chip sizes, we soon reach the moment, when we would need half an electron to move which is not possible. Networks are increasing in bandwidth, but as for instance key-note speaker Kleinrock pointed out, the limit is the speed of light. This gives a limit for the latency: the time it takes for a signal to reach its destination. So although the bandwidth available for the high-end is now getting into the Gbit/s, latency is preventing further development.
At SC2001 in Heidelberg, Dr. Jacek Marczyk, EASi Engineering GmbH, Alzenau, Germany, described their business which consists of about 400 engineers, providing services using stochastic modelling techiniques for solving problems within the automotive industry. His main thesis is that problems such as crashes, exhibit chaotic behaviour and therefore deterministic models are often inadequate. Using points to create functions can be dangarous because the points chosen may not be a true representation of the physics. He went on to say that Western culture prefers certainty to knowledge. The aim should be to run ensembles of the model, quantify the risks as a statistical distribution and then proceed to shift this distribution to provide the knowledge needed for re-engineering a better product, such as a car. (author: Chris Lazou).
At the SC 2001 Press Conference, Mr. Eric Schnepf from Fujitsu Siemens was proud to announce a freshly signed contract with the car manufacturer Audi, headquartered in Ingolstadt. Audi has selected Fujitsu Siemens to deliver the supercomputer hpcLine with 128 processors for its new strategic competence centre where stochastic crash simulations are being performed. After evaluating several RISC based and Intel based solutions, Audi's choice fell on Fujitsu's Intel based computing platform to simulate random crash scenarios. The system will be customised to meet Audi's specific needs for scientific and technical applications, and will consist of an integrated Linux cluster system based on 52 Pentium III dual processor nodes and 24 Pentium 4 nodes.
Professor of Computer Science at the University of Freiburg, Dr. Hans Burkhardt addressed the topic of high performance systems applied for serial image processing and database searching at SC 2001, the annual Supercomputing Conference, held in Germany. Especially the 3D data sets acquired from tomography scans and visual screenings in biology constitute an enormous challenge for real time visualisation and analysis. Two areas in particular were highlighted by the speaker who introduced some remarkable examples to the audience with respect to image database queries and visual navigation in natural environments. In comparison with human visualisation performance, the present state-of-the-art in hardware is still low-end peanuts, barely able to deliver an advanced search on the semantic level. Dr. Burkhardt predicted that only within a hundred years from now, a system surpassing teraflop performance might match the power of the human eye.
At the Heidelberg supercomputer conference, we had the opportunity to talk to Surjit Chana, the new Vice President of the High-Performance Computing Server Group at IBM. His message? IBM will keep delivering on its long-term road map. The company will stay in touch with large customers, and participate in new technology developments like The Grid, adding a few things of its own, especially in chip development. A somewhat boring message? Yes it is, but one that certainly will be reassuring for its customers.
At SC2001 in Heidelberg, Dr. Joerg Stadler, Marketing Manager, NEC European Supercomputers Systems, presented the NEC Supercomputer Road Map for the next four years. The bottom line message is that NEC is committed to both the vector highly integrated capability systems, the NEC SX5 series, and the scalable capacity servers using the Intel Itanium IA-64 chip set and its successors, McKinley and Madison. Stadler described some interesting developments in the pipe line for both streams and then went on to explain the recent agreement licensing Cray to sell NEC SX systems with exclusive rights for sales in the U.S.A. (author: Chris Lazou).
Last key note speaker at the SC 2001 Conference was Dr. Peter Zoller from the University of Innsbruck in Austria, who introduced himself as the exotic bird among the audience of computer experts and engineers since he is a physicist. His talk addressed Information Technology for the 21st century, in the knowledge that the technologies of the future are the basic research of today. In physics, quantum mechanics is a big revolution. The innovative concept may be used for a new paradigm in HPC also, which can be described as quantum computing to solve complex problems. Dr. Zoller however stressed that the fate of visionaries is that their predictions usually are either outperformed or not achieved.
At the Supercomputer Conference in Heidelberg, Leonard Kleinrock delivered the first key-note presentation. The Internet is fine but today keeps you behind your office desktop. When you get off to travel, to your home, to some customer office or to a conference, you become a computing nomad in the desert. You loose your fast access, your loose your technical support, and fall back to your stand-alone devices connected at a low speed, with small screens, small keyboards and lots of different batteries. The Internet is not ready for Nomadic Computing yet.
Dr. Horst Simon, Director at the National Energy Research Scientific Computing (NERSC) at the Lawrence Berkeley National Laboratory, in his keynote talk at SC 2001, shared some of his viewpoints with the audience on what he expected will be the major trends for the next five years in scientific computing. These developments include the rapid growth of micro-processors, the increasing use of open software; the optimised network bandwidth which will grow at an even faster rate than Moorḙ墳 Law has predicted; the somewhat paradoxical evolution towards aggregation, centralisation, and co-location; and the emergence of commodity products which will trigger the use of IP on everything that you can imagine.
To keep up to date with HPCN news in Europe, subscribe to Primeur weekly. Primeur Weekly is published per e-mail. Check out the subscription information for more details how to subscribe. You can find the back issues on the Primeur web site.
© 2001, Genias Benelux