rCUDA – Leveraging Low-Power Processors & InfiniBand Interconnects

Print Friendly, PDF & Email

In this video from the HPC Advisory Council Spain Conference, Federico Silla from the Technical University of Valencia presents: rCUDA – Towards Energy Efficiency in GPU Computing by Leveraging Low-Power Processors & InfiniBand Interconnects.

The use of GPUs to accelerate general-purpose scientific and engineering applications is mainstream nowadays, but their adoption in current high performance computing clusters is primarily impaired by the trend of including accelerators in all the nodes of a cluster, as this presents several drawbacks. In this talk, we introduce the rCUDA remote GPU virtualization framework, which has been shown to be the only one that supports the most recent CUDA versions, in addition to leveraging the InfiniBand interconnect for the sake of performance. Furthermore, we present the last developments within this framework, related with the use of low-power processors as well as enhanced job schedulers.”

Download the slides (PDF) or check out our HPCAC Spain Conference Video Gallery.