Victorian Partnership for Advanced Computing

Revolutionary advancements in technology now enable scientists, engineers and other professionals to perform computations, normally handled by the Central Processing Unit (CPU), on a Graphics Processing Unit (GPU).

Today's modern GPUs operate as parallel processors, which can be seen as the equivalent to several CPU's. This means that when performing computations on a GPU, as opposed to the CPU, researchers may find significant improvements in the speed of their jobs.

Through projects such as the 2008/09 Summer Internship, VPAC has researched and tested software to run on GPUs.

VPAC has now integrated four commodity class GPU machines and the Tesla C2050 demonstration unit 'Enrico' into fully functional nodes on Tango. The integration of these nodes will allow users at VPAC's member universities to trial GPU techniques first hand, with their own code written in the Compute Unified Device Architecture (CUDA).

Researchers interested in using VPAC's commodity class GPU nodes must have a VPAC HPC account which they can use to log into the clusters. If you do not have a VPAC HPC account, simply follow the link to apply for an account .

To access the Enrico demonstration unit provided by Xenon Systems please visit the Enrico GPU unit webpage for further information.

Globally, an increasing number of applications have been ported to GPGPU technologies over the previous year. Much of that work can be found at the nVidia CUDA Zone . If any VPAC users desire to compare an applications performance on a traditional cluster with Enrico, VPAC encourages you to acquire and load the CUDA-ised application, and run comparatively.

For further information on submitting jobs to VPACs GPU nodes visit the GPU user tutorials page.