Performance Evaluation of MPI Libraries on GPU-Enabled OpenPOWER Architectures: Early Experiences

Khorassani, Kawthar Shafie; Chu, C. Y. Cyrus; Subramoni, Hari; Panda, Dhabaleswar K.

doi:10.1007/978-3-030-34356-9_28

Cited by 16 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hanford et al [20] highlights shortcomings of current GPU communication benchmarks and shares experiences with tuning different MPI implementations. Khorassani et al [21] evaluates the performance of various MPI implementations on OpenPOWER systems. Chen et al [22] proposes compiler extensions to support GPU communication in the UPC programming model.…”

Section: Related Workmentioning

confidence: 99%

GPU-aware Communication with UCX in Parallel Programming Models: Charm++, MPI, and Python

Choi

Fink

White

et al. 2021

2021 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)

View full text Add to dashboard Cite

As an increasing number of leadership-class systems embrace GPU accelerators in the race towards exascale, efficient communication of GPU data is becoming one of the most critical components of high-performance computing. For developers of parallel programming models, implementing support for GPUaware communication using native APIs for GPUs such as CUDA can be a daunting task as it requires considerable effort with little guarantee of performance. In this work, we demonstrate the capability of the Unified Communication X (UCX) framework to compose a GPU-aware communication layer that serves multiple parallel programming models developed out of the Charm++ ecosystem, including MPI and Python: Charm++, Adaptive MPI (AMPI), and Charm4py. We demonstrate the performance impact of our designs with microbenchmarks adapted from the OSU benchmark suite, obtaining improvements in latency of up to 10.2x, 11.7x, and 17.4x in Charm++, AMPI, and Charm4py, respectively. We also observe increases in bandwidth of up to 9.6x in Charm++, 10x in AMPI, and 10.5x in Charm4py. We show the potential impact of our designs on real-world applications by evaluating weak and strong scaling performance of a proxy application that performs the Jacobi iterative method, improving the communication performance by up to 12.4x in Charm++, 12.8x in AMPI, and 19.7x in Charm4py.

show abstract

Section: Related Workmentioning

confidence: 99%