NVIDIA Deploys Tesla More Easily Thanks to HP

HP may have slammed the IT industry with the decision to change its business outlook, but it did not stop its current projects, such as the one that is set to benefit NVIDIA's supercomputing efforts.

GPU computing modules are something that have gained traction on the supercomputing segment, thanks to their high parallel processing prowess.

NVIDIA has a whole series of such products, dubbed Tesla, the same way AMD has the FireStream collection.

The former can now safely say that it has one more asset up its sleeve in the effort to further promote its modules, courtesy of HP.

What the latter did was create the GPU Starter Kit, essentially a pre-configured system, that puts at the disposal of researchers a full-featured GPU computing cluster.

The server carries the name of HP ProLiant SL390 G7 and has 24 Tesla M2070 GPUs, plus 16 CPUs and the NVIDIA CUDA 4.0 parallel computing software (offers GPU-accelerated libraries such as FFT, BLAS, LAPACK, RNG, and SPARSE, among other things).

13.5 teraflops is the peak performance attainable, while the price is of $99,000.

"Growing demand for GPU computing has fueled the need for a fully integrated, robust and affordable development platform that enables developers to easily create new, accelerated applications," said Sumit Gupta, manager of Tesla products at NVIDIA.

"As a result, NVIDIA and HP developed the GPU Starter Kit to remove some of the last hurdles to the mass adoption of GPU computing -- namely, the cost and time of system implementation."

Universities, government organizations and research departments should all be able to benefit from the acquisition of one such out-of-the-box cluster. HP even offers a set of third-party development tools at a discount.

The official press release that NVIDIA posted on its website contains all the information which those interested might wish to get a closer look at.