Multi-tenant virtual GPUs for optimising performance of a financial risk application

Javier Prades, Blesson Varghese, Carlos Reano, Federico Silla

Research output: Contribution to journalArticlepeer-review

9 Citations (Scopus)
369 Downloads (Pure)


Graphics Processing Units (GPUs) are becoming popular accelerators in modern High-Performance Computing (HPC) clusters. Installing GPUs on each node of the cluster is not efficient resulting in high costs and power consumption as well as underutilisation of the accelerator. The research reported in this paper is motivated towards the use of few physical GPUs by providing cluster nodes access to remote GPUs on-demand for a financial risk application. We hypothesise that sharing GPUs between several nodes, referred to as multi-tenancy, reduces the execution time and energy consumed by an application. Two data transfer modes between the CPU and the GPUs, namely concurrent and sequential, are explored. The key result from the experiments is that multi-tenancy with few physical GPUs using sequential data transfers lowers the execution time and the energy consumed, thereby improving the overall performance of the application.
Original languageEnglish
Pages (from-to)28-44
Number of pages17
JournalJournal of Parallel and Distributed Computing
Early online date17 Jun 2016
Publication statusPublished - Oct 2017


Dive into the research topics of 'Multi-tenant virtual GPUs for optimising performance of a financial risk application'. Together they form a unique fingerprint.

Cite this