June 21st, 2007
NVIDIA Unveils Tesla GPU and “Deskside Supercomputer”
By Michael Santo
Executive Editor, RealTechNews
NVIDIA today unveiled a line of high-performance processors. They’re still GPUs, but working in concert with a workstation they can bring the power of a supercomputer to workstations or servers.
The Tesla GPU (graphics processing unit) features 128 parallel processors and delivers up to 518 gigaflops of parallel computation. A gigaflop refers to the processing of a billion floating point operations per second. Nvidia envisions the Tesla being used in high-performance computing environments such as geosciences, molecular biology, or medical diagnostics.
Nvidia also will offer Tesla in a workstation, which it calls a Deskside Supercomputer, that includes two Tesla GPUs, attaches to a PC or workstation via a PCI-Express connection, and delivers up to 8 teraflops of processing power. A teraflop is the processing of a trillion floating point operations per second.
A Tesla Computing Server puts eight Tesla GPUs with 1,000 parallel processors into a 1U server rack. Source: InfoWorld
We Say: I actually expected AMD and ATI to come up with something like this. It’s a very impressive set of hardware, and it’s set to debut in August, with the Computing Server sampling in September.
John Corliss says:
Intell and AMD won’t do something like this when they can continue to dribble out minor improvements to their CPUs. It’s similar to the way that places in Europe have access to 40 mbs internet connections while people in the greedy, capitalistic United States only get 1 to 1.5 and have to pay more for 4.
June 21st, 2007 at 4:03 am
Dale says:
Keep in mind this isn’t really for general computing. Something similar was done with the folding@home client. http://folding.stanford.edu/FAQ-highperformance.html
This is for massively parallel processing.
June 21st, 2007 at 6:32 pm