pith. sign in

arxiv: 2410.00801 · v1 · pith:HEMCW3YWnew · submitted 2024-10-01 · 💻 cs.DC

Understanding Data Movement in AMD Multi-GPU Systems with Infinity Fabric

classification 💻 cs.DC
keywords gpuscommunicationmemorysystemsmulti-gpuapplicationsdataevaluation
0
0 comments X
read the original abstract

Modern GPU systems are constantly evolving to meet the needs of computing-intensive applications in scientific and machine learning domains. However, there is typically a gap between the hardware capacity and the achievable application performance. This work aims to provide a better understanding of the Infinity Fabric interconnects on AMD GPUs and CPUs. We propose a test and evaluation methodology for characterizing the performance of data movements on multi-GPU systems, stressing different communication options on AMD MI250X GPUs, including point-to-point and collective communication, and memory allocation strategies between GPUs, as well as the host CPU. In a single-node setup with four GPUs, we show that direct peer-to-peer memory accesses between GPUs and utilization of the RCCL library outperform MPI-based solutions in terms of memory/communication latency and bandwidth. Our test and evaluation method serves as a base for validating memory and communication strategies on a system and improving applications on AMD multi-GPU computing systems.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.