|
|
||
| 11 December 2000 |
EuroFlash no. 438
|
|
|
Advertisement
Primeur is the premier news service on HPCN and supercomputing in Europe. Primeur Weekly delivers the news each week in your e-mail box. Check out the Primeur web site for the Calendar, the Analysis section with background on the TOP500, the Monthly en Live! special issues, information on HPCN centres and industry.
|
||
NEC will deliver its 16-processor Azusa, which is Itanium based, to Bull that will package it with its own AIX5L operating system or Linux and sell it as part of its range of compute servers.
SGI and SuSE formed an alliance under which, SGI will make an undisclosed equity investment in SuSE and will work cooperatively on the development, deployment and support of the Linux operating system and infrastructure code.
The Research Institute for the Management of the Archives and Libraries at the University of Urbino in Italy joined NPACI. It will collaborate with NPACI partners on models for the long-term preservation and access to electronic records through the Data Intensive Computing DICE thrust area and XML technologies.
Cray will install a 32-processor Cray SV1 supercomputer system at the Warsaw University's Interdisciplinary Centre for Mathematical and Computational Modeling (ICM), Warsaw, Poland. The Cray SV1 system, to be installed this year in the ICM facility, will be used to enhance national weather forecasts for Poland and to support research in the mathematical, natural and computational sciences.
IBM and Telia, Scandanavia's largest telecommunications and internet service provider, announced tha Telia will install a combination of IBM mainframe and Shark storage technology, both running on Linux, to host and run their business and consumer Internet services operations across Scandanavia. Using this technology, Telia will also offer customers in the Nordic countries IP-VPN (Internet Protocol/Virtual Private Network) services.
Mega Co., Ltd., an independent Internet service provider (ISP) based in Tokyo, has adopted Turbolinux Cluster Server for its new clustered server system featuring total disk storage capacity of 1.5 terabytes and serving more than 100,000 corporate customers.
CARNet, the Croatian Academic and Research Network, has been connected to the European research backbone TEN-155 by a 34 Mbps link to Austria. As a result of other upgrades TEN-155 is now running at speeds of up to 622 Mbps.
In the late 1980's it became fashionable to give presentations with sound bites, such as, "supercomputers and the killer micro". The message was that cheap micros will replace the large scale supercomputers then in use, for solving the pressing problems of our time. Today the "killer Beowulf" has arrived; a cheap solution to supercomputing. A lot of time and money albeit in small parcels have been spent by "Geegs" to install clusters of Pentium, Alphas', Power 3 or other makes of commodity Chip, on a DIY (Do It Yourself) basis. These systems are mostly experimental relatively small, 32-64 nodes, cheap in capital outlay and in this author's view, are fraught with danger, especially in novice hands. "Buy (DIY) in haste, repent in leisure", as the Romans used to say.
Visiting Europe, Bob Bishop, SGI's Chairman and CEO, presented the results of the company's restructuring and refocusing. Back to its roots, SGI sees its single mission in supplying the technical and creative communities with high end and high-performance visualisation computers. The big data is connecting these markets. He additionally presented the MIPS and Intel IA-64 roadmap of the new servers and the cooperation with SUSE in Germany.
In this report we give an impression of the SC2000 Conference held in Dallas, Texas, USA, on 4-10 November. We discuss some trends and new developments as noted at the conference and the associated exhibition.
This year's keynote address was delivered by Steve Wallach, a supercomputing veteran, co-founder of Convex and for long years CTO for High Performance Computing at Hewlett Packard when it took over Convex. He is now a venture capitalist acting in the HPC field.
Traditionally, the technical program is mostly of high quality and this year this was at least true for the talks the author has attended. For the regular papers there were 2 parallel tracks of half hour talks that ran along with the Masterworks invited papers. As the field of HPCN has broaded over the years this was reflected in the session subjects.
Grids were one of the main themes of the SC2000 conference, apart from addressing them in panels, also a technical session was devoted to the subject. William Johnson from NASA Ames described NASA's Information Power Grid (IPG) in which they want to combine a large amount of services ranging from available HPC cycles to storage for massive datasets.
The interplay between MPI and OpenMP was another sign of the coming of age of OpenMP. In the cluster infrastructure session the first results of comparing Myrinet against Giganet's cLan were presented. In the applications sessions Bill Gropp from Argonne National Laboraroty reported about experiments with an unstructured mesh CFD code using PETSc, a parallel scientific library developed at ANL
The State-of-the Field talks at SC2000 were mostly enjoyable experiences. Four plenary, web-casted presentations were given in this track. The first one was by Thomas Sterling about the birth and development of the Beowulf cluster phenomenon. The second talk in this track was had a completely different nature. It was entitled ``A small dose of Infosec'' by the speaker Gene Spafford and addressed the field of computer security, or rather the lack of it. Margaret Wright from Bell Labs. was the third person featuring in this series. The title of her talk was ``Numbers, lots of numbers and insight too''. The last speaker in this track was J.C Browne from the University of Austin, Texas. He gave an overview of languages for Parallel and Distributed computing.
Although the exhibition floor was not noticeably larger that in the last few years, the amount of exhibits was certainly larger. The reason for the growing activity can be found in the large number of small startup companies that try their hand in cluster computing, grid computing, or what was dubbed Megacomputing in the Panels section. There was a variety of firms that sell preconfigured Beowulf clusters, the vast mojority based on Linux, some based on Window, and one (BTG) that offers a SparC-based cluster with the Solaris OS for Beowulf prices.
Locus Discovery a privately held computational chemogenomics biopharmaceutical company raised $42.75 million in a private round of equity financing through the issuance of preferred stock. y. Locus will use the funds to continue developing its revolutionary supercomputer-based approach to drug discovery, including continued advancement of the Company's drug development programs, establishment of a new research and development laboratory and office facility, and the hiring of a significant number of scientific and other personnel.
JCI Corporation today announced new storage service capabilities and the future direction of an engineering collaboration that was jointly announced by JCI and SGI (NYSE: SGI) on September 27, 2000. JCI Corporation plans to add 386 terabytes of SGI Total Performance storage solutions to its fiber optic Extranet by the end of 2002 to meet the demands for its FIRELINE services by entertainment creators and postproduction firms.
At SC 2001, to be held November 10-16, 2001, in Denver, a virtual counterpart will be organised called SC Global. It is global technical conference supported by emerging Access Grid technology.
Hewlett-Packard Company introduced a new pricing program, available initially for HP 9000 Superdome high-end UNIX servers. Referred to as partition pricing, this option allows service providers, enterprises and other Internet powerhouses to pay for incremental computing resources in alignment with forecasted growth. The program also offers an attractive entry price for the high-end HP Superdome solution and meets Internet data center needs for flexibility as well as for fast planning and implementation.
Unisys have signed a letter of intent for Dell to market a 32-processor Dell PowerEdge server based on the Unisys Cellular MultiProcessing (CMP) server architecture.
The Computational Science Resource Community (CSRC) debuted at SC2000. The CSRC is an interactive community-building too to house on-line learning materials, assignments, reviews, and people with a connection to the field of computational science.
Sun Microsystems introduced Sun Cluster 3.0, a clustering platforms for the service-driven network. Through automatic fault detection, recovery and redundancy, Sun Cluster 3.0, with the ability to cluster up to 8 nodes and scale up to 512 processors, ensures that mission-critical applications and services remain continuously available.
ITU Ventures, a venture capital firm that specifically focuses on financing businesses emerging from the nation's leading graduate universities, announced a seed investment in Simulant, Inc. Simulant has developed and is currently in beta release of a scalable distributed computing platform applicable to a broad class of both scientific and non-scientific problems. Simulant's distributed computing platform allows networked clusters of inexpensive personal computers to combine and duplicate the computational power of costly, high-performance supercomputers.
Envive Corporation will use Distributed Science's network of over 145,000 computers from 72,000 unique suppliers to create a worldwide Internet infrastructure for its performance testing and monitoring services.
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. The EuroFlash! and USflash! are published together with ESIS - European Supercomputing Information Service.
© 2000, Genias Benelux