Fujitsu Siemens Computers announces its new hpcLine solution using the latest Intel Xeon processors, codenamed Nocona. The new Intel Xeon processor based hpcLine compute node includes support for new technologies including Intel Extended Memory 64 Technology (EM64T), PCI Express and DDR 2 memory. With EM64T, users will be able to run both 32-bit and 64-bit applications so that they can use 64-bit computing for applications that take benefit of bigger memory addressing capabilities while continuing using 32-bit codes. Based on the Intel server chipset E7520, this new Intel Xeon hpcLine compute node features an 800MHz front-side bus.

End of July ZAM installed a new software level on its high-performance IBM Regatta cluster, named Jump, which specifically updated the microcode for the Federation Switch, the operating system components and the communication libraries. The latency was reduced by 40%, the transfer rate improved by a factor of 2.3. ZAM measured with its benchmark suite an improvement of 10% without any changes when using several nodes.

PathScale has announced an engineering initiative with Sun Microsystems and AMD to jointly tune Sun's Opteron-based systems. This engineering collaboration enables customers to achieve greater system throughput, setting a new standard for application-based performance results on Sun's AMD Opteron-based systems running Linux OS. With the PathScale EKO Compilers, the new AMD Opteron-based servers and workstations from Sun posted record setting benchmarks.

ClearSpeed Technology plc has successfully raised £11.1 million through the issue of new ordinary shares in relation to the proposed application for admission to AIM ("the Placing”). KBC Peel Hunt Ltd is nominated adviser and broker to the company.

Fujitsu Siemens Computers has launched PRIMERGY BX600 4-way Server Blade which is ideal for power-hungry data centres. The new blade will be available in August and is currently the most compact SCSI server blade worldwide. At a starting price of 8000 euro (plus VAT), the BX600 is the ideal platform for power-hungry enterprise applications due to its high performance and flexibility.

Verari Systems will offer BladeRack blade server systems based on the Intel Xeon processor running at 3.60 GHz with 800 MHz system bus and Intel Extended Memory 64 Technology (EM64T). Verari Systems' flexible and scalable platform-independent architecture ensures seamless incorporation of the latest in next generation high-performance technology.

The Ohio Supercomputer Center (OSC) has awarded a supercomputer cluster to the University of Toledo (UT) as part of its "Cluster Ohio” programme. Delivered on August 4 to UT's Department of Physics and Astronomy, the cluster was part of a larger system divided among institutions statewide.

The Ohio Supercomputer Center (OSC) has awarded a supercomputer cluster to Case Western Reserve University as part of OSC's "Cluster Ohio" programme. Delivered on June 28 to Case's physics department, the cluster was part of a larger system divided among several institutions statewide.

IBM has introduced new "scale-out" IBM eServer xSeries servers that include additional , high-performance features inspired by IBM mainframes and supercomputers.

HP continued its no. 1 position in worldwide server shipments for the ninth consecutive quarter, according to second quarter calendar year 2004 figures released by IDC. HP shipped nearly twice the number of servers than any other company during the quarter. HP also was the no. 1 server vendor in factory revenue and servers shipped worldwide for the x86, Linux and Windows server markets.

In the second quarter of 2004 Sun outperformed all vendors in revenue market share growth on a quarter-to-quarter (Q/Q) basis including All OS, UNIX+Linux and UNIX, as reported in the 2Q04 Worldwide Server Database report by Gartner Dataquest, released this week.

In this interview with EnterTheGrid - Primeur Magazine Mark Linesch, the new Chair of the Global Grid Forum (GGF), gave his views on the directions the Global Grid Forum should take and how he sees his role in the overall process. Mark Linesch is new to GGF but has a lot of experience with emerging technologies. "I was elected to represent the entire community not just one faction", he says, commenting on the wide variety of communities active in GGF. For instance research and commercial people. Communities from the US, Europe and Asia. Mark Linesch would like to see a transparent process with tangible results created in an environment where there is a diversity of ideas in terms of internationality, in terms of different types of workloads: business versus research.

The three largest public universities in the US state of Michigan — Michigan State University, the University of Michigan and Wayne State University — are creating a very high-performance network with the capacity to meet their emerging and demanding research needs. The project is especially critical for faculty whose research collaborations require the ability to transmit massive amounts of data across the network.

The NEES System Integration team has released NEESgrid 3.0, the final version of the software that allows earthquake researchers to integrate physical experimentation and model-based simulation, computational analysis, and improved testing and validation of increasingly complex and comprehensive analytical and numerical models.

The US National Center for Supercomputing Applications (NCSA) has joined the Globus Alliance, a consortium dedicated to the collaborative design, development, testing, and support of the open source Globus Toolkit. Globus provides key enabling software and services that let people share computing power, databases, and other tools securely online across corporate, institutional, and geographic boundaries without sacrificing local autonomy. It has been deployed broadly worldwide for both science and industry.

Could your home computer help cure Alzheimer's disease? Vijay Pande, PhD, assistant professor of chemistry and of structural biology at Stanford University, believes the answer may be yes. He's devised a way to identify potential drug compounds by using a network of more than 150,000 home computers and some innovative algorithms. He said the method accurately predicts how well molecules will bind to a given protein. Proteins are the ubiquitous workhorses of living systems and most diseases can be traced to protein malfunctions of one kind or another, so designing a compound that binds to a particular protein is an early step in drug development.

Virtual Compute Corporation will dedicate a portion of their supercomputing infrastructure to the biomedical segments of the "On-Demand" HPC industry. The new vCompute Bioinformatics Group has a very interesting approach to medical research by applying advanced supercomputing and data mining methodologies.

On August 26th, the climateprediction.net project began using the Berkeley Open Infrastructure for Network Computing (BOINC) software platform, developed by scientists at the University of California in Berkeley.

Orion Multisystems inked agreements with Wolfram Research and The BioTeam to implement leading productivity software in Orion's new 96-node Deskside and 12-node Desktop Cluster high performance Workstations for technical computing.