The maker of jolly expensive printer ink is getting all excited about shoving Nvida GPUs where the sun does not shine – the heart of a data centre.
HPC experts are enthusiastic supporters of GPUs for computing tasks because, using parallel processing, they can get much more performance without needing more electricity.
Writing in an Nvidia bog, Ed Turkel, marketing manager for HPC for HP said that the HPC segment has an “endless thirst for performance” and GPUs were a disruptive force, for the status quo.
He claimed that for boffins enhanced performance allows for faster innovation that will result in the kinds of discoveries.
Nvidia claims its Tesla M2090 is the world’s fastest parallel processing chip. It can have 512 cores and can make software go ten times faster than an ordinary CPU.
HP said that the Tesla M2090 GPU will be available on HP’s new ProLiant SL390 G7 4U server, which Nvidia officials said was built specifically for hybrid computing environments.
Turkel, said that the Tsubame 2.0 supercomputer in Japan, which is the world’s fourth-fastest supercomputer, is typical of what GPU computing can do. That system is based on 1,408 ProLiant SL390s G7 servers, and delivers similar peak performance as Cray’s XT5 Jaguar system at the Oak Ridge National Laboratory.
However, Tsubame 2.0 has 92 percent fewer servers and consumes one-seventeenth of the power than Jaguar. Needless to say the Jaguar is run on AMD chips.