 |
| News digest 29 June 2007 |
|
|
 |
| |
| Start |
|
|
| PrimeurLive! on ISC2007 in Dresden |
|
This year we will again report on the ISC supercomputer conference in Dresden. ISC2007 is the major supercomputing in Europe.
We did publish the following issues:
In addition a special has been published in:
|
|
|
|
| ISC'07 Reaches Record Figures |
|
Europe's supercomputer conference ISC'07 in Dresden that finished earlier today, did attract 1213 participants from 45 different countries. The exhibition did see 85 exhibitors. This makes it the largest supercomputing event on this side of the Atlantic. Missed the event this year? Mark your agenda for next year's ISC'08 from 17 -20 June 2008, again in Dresden, Germany.
Read further... |
|
|
|
Advertisement
|
|
Advertisement
|
|
| |
| TOP500 |
|
|
| Tera-10' 42.9 teraflops drives Europe's HPC |
|
Tera-10, the supercomputer installed in by Bull and connected by 3 rails of Quadrics QsNetII interconnect, has become the fastest system in Europe and the 5th in the world. Tera-10 is the second stage of the TERA project which aim to install at CEA/DAM a computing power of 100 sustained teraflops (half a petaflops peak) in 2010. The announcement was made in the 27th TOP500 List of Supercomputers which went published at the Dresden ISC2006 conference. Using the industry-standard LINPACK benchmark, the Tera10 system attained a sustained performance of 42.9 Teraflops, eclipsing the top mark of 20.5 Teraflops for the MareNostrum system, in Spain.
Read further... |
|
|
|
| Appro Cluster solution ranked the 19th fastest supercomputer in the world |
|
The "Atlas" Appro XtremeServer Clusters at the Lawrence Livermore National Laboratory (LLN) has been officially ranked the 19th fastest in the world according to the current TOP500 list of supercomputers. The APPRO XtremeServer Cluster, "Atlas" at Lawrence Livermore National Laboratory, reached a Linpack benchmark performance 36.580 Rmax TFlop/s (teraflops) or trillions of floating point calculations per second.
Read further... |
|
|
|
Advertisement
|
|
| |
| Hardware |
|
|
| Europe unites to foster HPC technologies with the launch of ParMA, an ITEA 2 project to exploit fully the power of multi-core architectures |
|
Leading European HPC players from France, Germany, Spain and the United Kingdom, have joined forces to launch the ParMA (Parallel Programming for Multi-core Architectures) consortium. The key objective of the new consortium is to help the HPC community to benefit fully from the processor race while developing innovative, flexible and open technologies for taking full advantage of the multi-core architectures.
Read further... |
|
|
|
| |
| Applications |
|
|
| Goethe's view on high performance computing in the bank |
|
At the ISC'07 press conference, Dr. Daniel Egloff, Head Financial Computing of the Zurich Cantonal Bank, shed a light on the use of high performance computing in the financial service sector. The financial world is a universe full of dangerous risks. The question whether high performance computing can help increase a bank's profitability and enhance its stability in times of economical crisis, was left open to the audience by the speaker who cited Goethe to make his point.
Read further... |
|
|
|
| |
| The Grid |
|
|
| GARUDA, the National Grid Initiative of India |
|
At ISC2007 in Dresden, Dr. Rajendra Joshi, Group Coordinator Bioinformatics of the Centre for Development of Advanced Computing in India did present GARUDA, the Indian National Grid Initiative, as part of the European & Asian Grid Summit in Dresden.
Read further... |
|
|
|
| |
| Company news |
|
|
| Quadrics announces new pricing for its 10 Gigabit/s Ethernet Switches-QsTen G |
|
The new price is a further step towards the adoption of 10Gbps-Ethernet in HPC environments. Attractive pricing together with the ubiquitous nature of Ethernet will catalyse the adoption of this standard in server clusters and storage servers.
Read further... |
|
|
|
| TotalView Technologies announces TotalView 8.2 |
|
TotalView Technologies, a provider of scaleable debugging and analysis software solutions for the multi-core age, has made available TotalView Debugger 8.2. This updated version of TotalView provides new features and support for important new platforms, allowing a greater range of users to benefit from the product's capabilities.
Read further... |
|
|
|
| Microsoft technical fellow keynotes on the reinvention of computing |
|
Fresh off the revelation that Microsoft high-performance computing (HPC) technologies are being widely adopted by customers, software and hardware partners, a technical leader from Microsoft Corporation discussed how the foundations of computer science and engineering must be reinvented to deal with the mass-market adoption of processors with many computing cores.
Read further... |
|
|
|
| QLogic leads HPC networking industry with over 40,000 DDR InfiniBand switch ports shipped |
|
QLogic Corp., an expert in networking for storage and high performance computing, is leading the industry by passing the 40,000 mark for DDR InfiniBand ports shipped. At ISC, QLogic is showcasing its end-to-end portfolio of Double Data Rate (DDR) InfiniBand networking products. QLogic has been delivering a full offering of DDR products for over a year and demonstrated this hardened end-to-end configuration at the International Supercomputing Conference this week.
Read further... |
|
|
|
| Bull and Platform Computing announce Worldwide Cooperation Agreement to optimize resource management in HPC clusters |
|
Bull and Platform Computing have signed a worldwide agreement for the availability of Platform LSF on Bull's NovaScale High Performance Computing (HPC) clusters. Under the terms of this partnership, Platform LSF is available on both Intel Xeon based Bull NovaScale Cluster 400 and Intel Itanium based Bull NovaScale Cluster 3000. The Platform LSF family of products is fully integrated with Bull Advanced Software (BAS), to help users manage and optimize complex IT environments delivering higher IT efficiency.
Read further... |
|
|
|
| HPC Systems to become Intel Cluster Ready Certified using Platform Open Cluster Stack |
|
Platform Computing will enable HPC Systems to become Intel Cluster Ready through Platform Open Cluster Stack (OCS). Intel Cluster Ready is a new programme and technology designed to simplify the development, purchase, deployment, usage and management of clustered computer systems. Platform OCS is the first commercial software system used to create certified Intel Cluster Ready systems that enable the consistent delivery of scale-out application clusters. HPC Systems, a Japanese integrated hardware vendor (IHV), will bundle Platform OCS and Platform LSF with their cluster offerings.
Read further... |
|
|
