Multi-Tenant Virtual GPUs for Optimising Performance of a Financial Risk Application

TL;DR

This paper explores multi-tenant GPU sharing in HPC clusters to improve performance and energy efficiency for financial risk applications, demonstrating that sequential data transfers with shared GPUs reduce execution time and energy use.

Contribution

It introduces a multi-tenant GPU approach with sequential data transfers, optimizing resource utilization and performance for HPC applications.

Findings

Multi-tenancy with shared GPUs reduces execution time.

Sequential data transfers improve energy efficiency.

Shared GPU approach enhances overall application performance.

Abstract

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.