Nvidia Reveals Tesla Solutions for High-Performance Computing

by Anton Shilov
06/22/2007 | 02:33 PM

Nvidia Corp., a leading developer of graphics processors and chipsets for computers, announced this week a series of products aimed at high-performance computing markets. The new Tesla processing blocks will feature several GeForce 8800-series graphics processing units (GPUs) that are capable of rapidly computing specifically made programs.

A number of simulation tasks, including, but not limited to, fields like geosciences, molecular biology, medical diagnostics and financial calculations, require tremendous amounts of computing power to be processed. But given that modern central processing units sometimes cannot provide enough computational resources for such tasks, scientists have to either utilize huge power-consuming clusters with hundreds of such chips, or reprogram their applications for some other types of processing units. Given that the nature of graphics processors relies on simultaneous multi-threading, GPUs are very likely candidates for scientific computations.

Under the Nvidia Tesla brand the company will offer a family of GPU computing products that will be able to provide high-performance computer power at a relatively low price.

Currently Nvidia Tesla family of products, based on the GeForce 8800 chips, consist of the following:

• Nvidia Tesla GPU Computing Processor, a dedicated computing board that is installed into a typical PC or workstation.
• Nvidia Tesla Deskside Supercomputer, a scalable computing system that includes two Nvidia Tesla GPUs and attaches to a PC or workstation through an industry-standard PCI-Express connection. With multiple deskside systems, a standard PC or workstation is transformed into a personal supercomputer, delivering up to 8 teraflops of compute power to the desktop.
• Nvidia Tesla GPU Computing Server, a 1U server housing up to eight Nvidia Tesla GPUs, containing 1024 parallel processors that add teraflops of parallel processing to clusters. The Tesla GPU Server is the first server system of its kind to bring GPU computing to the datacenter.

Nvidia did not announce pricing or availability timeframes of Tesla-series products.