informa
/
1 MIN READ
News

Nvidia Offers GPU System For High-Performance Computing

The preassembled system is aimed at high-performance computing environments used in scientific research, the oil and gas industry, and the financial industry.
Nvidia on Monday introduced a preassembled computer powered by graphics processors that the company claims offer a less-expensive alternative to CPU-based systems used in high-performance computing environments.

The Tesla S1070 1U GPU computer uses substantially less power and provides much higher performance than CPU-only systems, according to Nvidia. The company has partnered with a number of computer makers in making the systems, include AMAX, Appro, CADNetwork, Colfax International, and Cray.

Each Tesla S1070 system includes up to four GPUs. The computers are configured into a cluster that can operate on its own or in conjunction with CPU-based systems. A pre-configured cluster consists of four S1070s delivering 16 teraflops. The cluster includes host servers, InfiniBand switches, and cabling.

The corporate and investment banking division of French banking company BNP Paribas has replaced 500 traditional CPU cores consuming 25 kilowatts of power with a smaller CPU cluster with two Tesla systems that consume a total of 2 kilowatts, according to Nvidia.

The new system is being aimed at high-performance computing environments used in scientific research, the oil and gas industry, and the financial industry. However, Nvidia claims the systems also can be used in corporate data centers.

Other tech companies offering Tesla pre-configured clusters include Dep, FluiDyna, HPC Technologies, Inspur, JRTI, Megware, Microway, NetWeb, PCPC, Penguin, Silicon Mechanics, Sprinx, T-Platforms, Viglen, and Xenon.

In November, Nvidia launched an HPC design for workstations using the company's Tesla graphics processors. The design used up to four Tesla GPUs, each contained on a card that slips into a PCI Express slot.


Learn more about all the latest products and technologies at TechWeb's Interop Las Vegas, May 17-21. Join us (registration required).

Editor's Choice
Samuel Greengard, Contributing Reporter
Cynthia Harvey, Freelance Journalist, InformationWeek
Carrie Pallardy, Contributing Reporter
John Edwards, Technology Journalist & Author
Astrid Gobardhan, Data Privacy Officer, VFS Global
Sara Peters, Editor-in-Chief, InformationWeek / Network Computing