You can rent the high-performance computing resources required to see into the universe, parse the composition of matter or analyse financial risk, thanks to HP. The low-cost computing revolution means that even modest-sized businesses can harness the kind of power once reserved for government agencies and aerospace giants.
The cost reductions are made possible, in part, by HP’s use of open standards and industry-standard servers, and by the declining cost of Internet bandwidth.
In a typical enterprise, an estimated 70 to 90 percent of data centre resources are underutilised. To address this problem, HP is also bringing virtualisation to its customers, allowing users to automatically grow and shrink resources based on application requirements.
“High-performance computing is at the heart of today’s scientific and technological progress,” says Martin Walker, HP's scientific research manager for Europe, the Middle East and Africa.
HP leads the high-performance computing (HPC) market in revenue with a 34 percent market share, according to first quarter 2005 figures released by IDC. 1
HP has opened the first open source utility performance centre in Europe, the HPC1, which is designed for prototyping and deploying complex IT solutions that work with HP's extensive product and service portfolio.
“We are democratizing access to the supercomputer,” explains Philippe Devins, the project’s leader who is also in charge of HPC sales for HP France.
The centre’s first customers have been oil, aerospace and automotive multinationals, as well as private and public research laboratories. Rick Owen, who is responsible for oil and gas simulation at Schlumberger, says that HPC1 was “a natural choice as the most viable, flexible, available and, by far, the most affordable solution…the return on investment was immediate.”
Philippe Devins believes that, within a year or two, medium-sized and even small businesses will form the bulk of his business at HPC1. The reason is cost. It’s a lot cheaper to rent a supercomputer for a few days or weeks than to own it yourself.
How would you like to triple your supercomputing performance, double your capacity – and pay less? If you are running an HP supercomputing cluster, you can – by purchasing an HP StorageWorks Scalable File Share (HP SFS), a Linux cluster file server.
The HP Unified Cluster Portfolio is a modular package of hardware, software and services that helps customers better compute, manage, store, access and visualise the large volumes of data created by the computational analysis used in such highly complex areas as human genomics and the development of nano-scale electronic devices.
"HP SFS delivers the high performance shared storage that we require to maximise the throughput in our HP supercomputing cluster," said Dr. Wilfried Juling, professor at the Faculty of Computer Science and director of the Scientific Supercomputing Center Karlsruhe in Germany.
"We depend on this supercomputing power to accelerate the industrial and theoretical research we do for the universities of the state of Baden-Württemberg and our industrial partners Porsche AG and T-Systems, a leading German provider of computing services,” he adds. “We plan to grow this cluster to 11 teraflops in 2006."
Another key part of the HP Unified Cluster Portfolio is the HP Scalable Visualisation Array, which allows scientists and engineers to visualise their designs as they exist in real life, with photo-realistic image quality. Driving a high-resolution display wall of 100 million pixels costs half as much as competitive proprietary products.
“The best way for humans to understand very large amounts of data is to present that data in visual form,” explains Walker.
Grid computing – the linking of PCs and servers via the Internet – is a powerful but cost-effective alternative to a supercomputer.
These virtual supercomputers, which can quickly process vast amounts of information, have helped produce breakthroughs in meteorology, physics, medicine and other fields.
“HP’s longer-term goal is to bring the benefits of grid computing to the corporate world,” explains Martin Walker. “Investment banks and securities firms,” he notes, “are putting grid computing to work for advanced risk management.”
HP is using industry standards to grid-enable HP products from servers to the most powerful storage arrays and supercomputers. Since one HP strength is creating software for managing complex environments, HP is designing software that helps enterprises manage grid infrastructures.
Grid computing will ultimately let enterprises use their IT resources more efficiently by allowing collaboration within a company on any kind of computing task – regardless of size or duration – making it possible to set up secure IT projects between geographically disparate units, and even across organisational boundaries.
“One day, grid computing will be as simple as an electric utility is today,” says Walker.
HP is the first and only commercial member of the operational grid for the Large Hadron Collider (LHC) – the world's largest scientific instrument – at CERN, the European Laboratory for Particle Physics. When it becomes operational in 2007, the LHC will collide tiny fragments of matter head-on to explore the fundamental laws of nature in intricate detail, producing vast quantities of data that will be distributed and processed on the LHC grid.
| (1) |
Worldwide Technical Server 1Q05 Vendor Shares, IDC.
|
Printable version
