|
|
||
|
20 June 2002 Live issue from the Heidelberg Supercomputer Conference, ISC2002, June 2002Primeur Live! is published during major Supercomputing and Grid events in Europe
Advertisement
Advertisement
The Heidelberg Supercomputer Conference (ISC2002) is the main supercomputing event in Europe. This year we publish two live issues from the conference:
Advertisement
|
Advertisement
More than 350 participants are attending the annual International Supercomputer Conference these days. Again more visitors at this 17th edition than last year, proving that recent developments in high performance and grid computing are stimulating colleagues from all over the world to join each other in Heidelberg to exchange new ideas. The fabulous results of the Earth Simulator in the Top500 probably have enhanced interest. The conference opened today with a traditional warm welcome to all HPC veterans and expert newcomers by Professor Dr. Hans Meuer from the University of Mannheim.
Jens Reich, biologist at Max-Delbrück Center of Molecular Medicine in Berlin, probably held the most philosophical and ethical related talk among the presentations at ISC 2002. Sketching the roots, history, and political implications of man's ultimate dream to create the perfect human race, Dr. Reich acknowledged possible useful developments in future genetic engineering but doubted the need to generate a perfect human being in what might turn out to be an individual existence in a not all that perfect society.
Monika Henziger from Google Inc. was invited to give the first key note presentation at ISC2002. She addressed the issues which have to be considered allowing users to efficiently and quickly find on the Web what they are searching for. To dig up relevant user information out of the two to ten billion pages on the Web, that are doubling every eight months, Google relies on smart algorithms and parallelism. Sounds like a typical job for powerful supercomputing.
This year again Ministerialrat Bernd Reuse, Federal Ministry of Education and Research (BMBF), gave an overview of the actual funding activities of BMBF in this field.
Songnian Zhou from Platform, Earl Joseph from IDC and Martin Walker from HP, discussed their views on the state-of-the art in Grid computing and the directions it will take.
The High-Performance Computing Center North (HPC2N), located at Umeå University, Sweden, has built a new HPC Linux cluster with a peak performance of 800 GigaFLOPS/s. This high performance is achieved using 240 AMD Athlon MP2000+ processors in 120 dual nodes and interconnected with the innovative WulfKit3, a high-speed interconnect system provided by Dolphin Interconnect Solutions that features a three-dimensional torus distributed switch topology.
DataDirect Networks, a provider of high performance storage networking appliances, is presenting 2Gb appliance-based SAN solutions and showcasing their Silicon Storage Appliances to attendees of the International Supercomputing Conference 2002 from June 19-22 at the Heidelberg Convention Center. Silicon Storage Appliances make it easy for the scientific community to deploy and scale high performance storage area network infrastructures in a cost-effective, modular approach to support supercomputing uses including the construction of terascale and distributed computational grids.
A new number one in the world, the Japanese Earth Simulator, a new number one in Europe, the HP AlphaServer SC ES45/1 GHz at CEA in France, a new company with the most machines in the list (HP/Compaq). A lot has happened in the world of supercomputing since the last TOP500 list was released in November last year. This can be seen from the latest list of the world's moste powerful supercomputers, released this morning at ISC2002 in Heidelberg.
With 64 machines in the top500 and over 20 Tflop/s of installed Linpack performance, Germany is still the most important European supercomputer country. Despite the fact that the first position in Europe is now belonging to a supercomputer installed in France. Impressive as 20 Tflop/s sounds, it is much less than the performance of the Earth Simulator machine that holds the world's first position.
The presentation of the Top500 June 2002 opened the way to analyse the German situation, as since the beginning of this year several new systems have been installed in research and academia, as well as extended Leibniz-Rechenzentrum in Munich.
In what has become a much-anticipated event in the world of high-performance computing, the 19th edition of the "TOP500" list of the world's fastest supercomputers was released today. The recently installed Earth Simulator supercomputer at the Earth Simulator Center in Yokohama, Japan, is, as expected, the clear new number 1. Its performance of 35.86 Tflop/s (trillions of calculations per second) running the Linpack benchmark is almost five times higher than the performance of the now no. 2 IBM ASCI White system at Lawrence Livermore National Laboratory (7.2 Tflop/s). This powerful leapfrogging to the top by a system so much faster than the previous top system is unparalleled in the history of the TOP500.
Professor Dr. Hans Meuer from the University of Mannheim presented a new benchmark initiative for distributed computer systems at ISC2002 today. The project is called IPACS, which stands for Integrated Performance Analysis of Computer Systems.
To keep up to date with Supercomputing and Gird Computing news in Europe, subscribe to Primeur/EnterTheGrid 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