|
|
||
| 13 November 2000 |
EuroFlash no. 434
|
|
|
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.
|
||
A general course on scientific supercomputing will by held in Karlsruhe, Germany, in February 2001. At the Universitaet Karlsruhe and the Forschungszentrum Karlsruhe an IBM RS/6000 SP and a Siemens/Fujitsu VPP300-5000 with totally 400 Gflop/s are installed. At LRZ Munich a Hitachi SR8000 with more than 1 Tflop/s is available for German scientists. Students will get acquainted with the usage of such supercomputers.
eGrid - the European Grid Forum, has combined with Grid Forum, a primarily U.S.-based organisation of individuals developing, deploying, or using grid technologies, and leaders from Asia Pacific to create Global Grid Forum (GGF)
TRAIN-IT has been restarted as an Accompanying Measure in the IST Programme, after finishing its very successful piloting phase.
PeopleSoft has embedded mathematical and statistical software components from NAG's C Library into PeopleSoft 8 Enterprise Performance Management products.
Fakespace Systems opened its first European development office, and appointed industry veteran Richard Cashmore, as Business Development Representative. The new office, located in Leicestershire, United Kingdom (UK), will enable Fakespace Systems to better support its current customers and aggressively expand its sales and marketing programs in key European markets.
Veritas DGC (VDGC), has installed their third NEC SX-5 Series parallel vector supercomputer. The newest VDGC installation is located at their geophysical processing facility in London. The Company recently installed SX-5 supercomputers at both their Houston and Singapore centres. VDGC currently operates 21 seismic data processing centres in major oil and gas markets world wide.
SuSE Linux and IBM bring DB2 Universal Database to Linux SuSE Linux Database Server provides companies with a basis for their e-business that expands with their growing demands and success.
HLRS in Stuttgart, organises a workshop on the use of the Hitachi SR3800 supercomputer on November 21st. The SR/800 in Stuttgart has a peak performance of 128 Gflop/s and 128 Gbyte of memory. The machine has 16 SMP-nodes, each with 8 processors. The SMP nodes can be used as a pseudo-vector unit with 8 Glop/s performance.
SuSE Linux announced the general availability of SuSE Linux Enterprise Server for S/390. SuSE's Linux distribution for the mainframe allows enterprise customers to scale Linux from PC clients to host systems.
This year's fall meeting of the different Fujitsu Siemens high-end computer users took place in the Munich facilities in Neu-Perlach South from October 9th to October 11, 2000. There, the Fujitsu International Supercomputer Users (FISUM) met the 9th, the SAVE Working Group Scientific Computing the 13th and the VPP Special Interest Group the 8th time. Users from all over the world, Japan and Australia as well as Europe, came together to discuss their experiences and the requirements with Fujitsu Siemens representatives too. This special report will cover first experiences with the Intel Pentium-based hpcLine, new developments, the porting of application packages on this architecture, virtual reality on hpcLine and some remarks from vector computer centers.
Parabon Computation announced general availability of its commercial distributed computing platform, Frontier. Any organization can now benefit from distributed computing via the Internet, the company claims. Shipment of Frontier will begin on November 15, 2000.
Cray unveiled the Cray SV1ex supercomputer product line at SC2000, the annual high-performance computing conference. The new line strongly enhances the performance and price/performance of the current Cray SV1 series. The air-cooled Cray SV1ex vector systems are slated for availability in the first half of 2001, at U.S. list pricing from $700,000.
High-performance projection technologies Christie Digital System did present a technology demonstration of the world's first stereoscopic 3-D digital projector, based on 3-chip DLP technology, at the Supercomputing Show in Dallas. Sponsored by Fakespace the demonstration will feature a single large-screen display in regular show-level lighting with interactive applications.
Many tried, but to date all efforts failed to define benchmarks that go beyond the Linpack used in the TOP500 and have wide acceptance by users. Eurobench, Parbench, the NAS kernels, and many others. But, nevertheless, the HPC User Forum organised by IDC, representing leading supercomputer users in government, industry and academia, reported progress on a plan to create better performance tests for this most powerful class of computers. Speaking at the SC2000 supercomputing conference in Dallas, HPC User Forum officials said improved tests are needed to advance scientific research, industrial engineering and classified government work, all of which rely heavily on supercomputers. It would be too easy to predict that also this very US oriented benchmark initiative will fail, but the approach, identifying a set of "representative" benchmarks is not different from earlier attempts. Supercomputer architectures from different vendors evolving into the same direction, could, however, help this initiative.
One of the foundations of the Grid is to connect resources from all kinds of providers, running all kinds of software. Most Grid initiatives use Globus services, or for instance in Germany Unicore, to provide a connectivity layer. With NPI - the New Productivity Initiative - this could change. Eleven companies have taken the initiative to define and open standard API for distributed resource management. When it is a success, it can accelerate the take-up of Grid services in commercial settings.
Synopsys announced that Cray has successfully taped out an eight million-gate, 450 MHz, vector processor ASIC using Synopsys' Physical Compiler. The design was implemented using a copper-based 0.12-micron process technology. The chip, code named Processor Vector Cache (PVC), is being used in the new Cray SV1e scalable-vector supercomputer, which was announced today at the IEEE Supercomputing 2000 conference in Dallas, Texas.
NASA has acquired Platform's LSF MultiCluster software to manage the combined scientific and engineering computing workloads at the Glenn Research Centre in Cleveland, OH and the Langley Research Centre in Hampton, VA. LSF MultiCluster extends the reach of Platform's market leading Load Sharing Facility (LSF) distributed resource management (DRM) capabilities by coordinating the scheduling of the computing resources between the NASA centres in a fail safe, lights-out, peer-to-peer operational relationship, without requiring the addition of new hardware and without having to learn or implement any new job submission language or queuing system syntax.
API NetWorks introduced the CS20 at SC2000, the world's most powerful server in a 1U package and the company's most advanced platform for computer clustering. The CS20 is a rack-optimized, 1U, dual-processor Alpha server and gives the high-performance computing (HPC) and Internet community the ability to scale clusters into state-of-the-art supercomputer facilities, providing the most economic solution with the minimum computer room footprint.
Duke University researchers will use an IBM SP supercomputer to create models of the heart that they hope will lead to uncovering causes and developing treatments for life-threatening heart conditions. Dr. John Pormann and the Electro-physiology research team at Duke University are creating accurate and complex modeling of electrical currents flowing through the heart and nerve tissue.
At the Supercomputing2000 conference this week in Dallas, Texas, Entropia CTO Andrew Chien announced this morning that his company has signed agreements with both the National Computational Science Alliance ("Alliance") and the National Partnership for Advanced Computational Infrastructure (NPACI). As part of the agreements, Entropia, an Internet and enterprise computing services company, will donate an unprecedented 200 million CPU hours to the PACI program, creating the largest computing platform ever offered to its national academic user community. The Alliance and NPACI are both part of the National Science Foundation's Partnerships for Advanced Computational Infrastructure Program.
Linux provider Atipa Corporation is now offering prospective customers free on-line testing of its new Ascendance Beowulf clusters.
SGI launched the Total Performance 9400 (TP9400) RAID storage array, a full-fiber storage solution that offers the highest throughput RAID storage subsystem in its class, the company says.
Many people who attend conferences say they go for the networking. But the 5,000 computing and networking experts at the SC2000 conference did have access to one of largest, most complex communications networks in the USA offering a combined capacity more than 196,000 times faster than a typical residential Internet connection and 200 times as fast as the connections used by many universities.
Saudi Aramco, a leading global petroleum company, has completed the installation of 92 node IBM SP supercomputer. It is based at the company's headquarters in Dhahran, Saudi Arabia.
The College of William and Mary installed one of the largest academic computing clusters world wide. The "SciClone" HPC cluster is actually a grid of four sub-clusters that consists of 160 processors and offers a theoretical peak performance of 115 Gflop/s. The grid-like environment will allow faculty and students to research the issues facing users of larger distributed systems utilizing the Internet.
Sun recently purchased additional LSF server licenses for its Systems Product Group, which designs a wide range of computer systems including enterprise appliances, desktop workstations, workgroup, mid-range and high-end servers, making it one of the largest users of Platform's LSF technology.
The Army Research Laboratory (ARL) Major Shared Resource Center (MSRC) has installed a 256-processor SGI Origin 3800 system as part of the Department of Defense (DoD) High Performance Computing Modernization Program (HPCMP). The supercomputer is the largest SGI Origin 3000 series shared-memory system delivered to date.
PV-WAVE is a family of software products that let engineers, scientists, researchers and business analysts understand large and complex datasets using a wide variety of visual techniques. PV-WAVE's interactive, array-based scripting language allows users to rapidly develop platform-neutral applications for manipulating, analyzing and visualizing these datasets to detect and display trends, patterns and anomalies. The software includes hundreds of sophisticated mathematical and statistical functions from the company's IMSL . C Numerical Library (CNL), as well as image processing, signal processing and general mapping functionality.
Raytheon Company will SGI Origin 3000 supercomputers at the Geophysical Fluid Dynamics Laboratory (GFDL) in Princeton, N. J. The upgrade is expected to improve the US climate prediction and weather forecasting capabilities. Eight 128-processor Origin 3800 systems will form a large-scale cluster and two 64-processor Origin 3800 systems will serve as an analysis cluster. The smaller analysis cluster will also serve GFDL's data archive that is stored in three robotic tape libraries and is expected to reach 2 Petabyte in size by September 2003.
Lawson, Inc., the Japanese convenience store chain, will run Linux on the IBM eServer xSeries to allow customers to download music, movies and other Web content from multimedia terminals throughout its stores. The terminals also will provide timely information collected from Internet feeds and communications satellites. Customers already can book airline reservations and order concert tickets from the existing terminals. Armonk will install more than 15,000 IBM eServer xSeries servers, running Linux, making it one of the largest IBM Linux-based installations to date.
olphin Interconnect announced the world's largest custom supercomputer cluster based on the Advanced Micro Devices (AMD) Athlon processor. The supercomputer physically comprises four standard racks containing 132 processor units in a parallel cluster architecture interconnected with high-speed WulfKit Scalable Coherent Interface (SCI) technology supplied by Dolphin Interconnect. The system, already christened "SAMPSON" will be installed at the Bartol Research Institute at the University of Delaware and is expected to rank among the world's top 100 supercomputers once its performance is benchmarked.
IBM's WebSphere Commerce Suite Start version 4.1 is now available on Linux. It can be used to provide customers the functionality to buy and sell products and services in a personalised manner on the open Linux platform, while extending their e-commerce applications to any server.
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