Performance Modeling of Streaming Kernels and Sparse Matrix-Vector Multiplication on A64FX

Alappat C, Laukemann J, Gruber T, Hager G, Wellein G, Meyer N, Wettig T (2020)

Publication Type: Conference contribution

Publication year: 2020

Publisher: Institute of Electrical and Electronics Engineers Inc.

Pages Range: 1-7

Conference Proceedings Title: Proceedings of PMBS 2020: Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems

ISBN: 9780738110486

DOI: 10.1109/PMBS51919.2020.00006

Abstract

The A64FX CPU powers the current #1 supercomputer on the Top500 list. Although it is a traditional cache-based multicore processor, its peak performance and memory bandwidth rival accelerator devices. Generating efficient code for such a new architecture requires a good understanding of its performance features. Using these features, we construct the Execution-Cache-Memory (ECM) performance model for the A64FX processor in the FX700 supercomputer and validate it using streaming loops. We also identify architectural peculiarities and derive optimization hints. Applying the ECM model to sparse matrix-vector multiplication (SpMV), we motivate why the CRS matrix storage format is inappropriate and how the SELL-C-σ format with suitable code optimizations can achieve bandwidth saturation for SpMV.

Authors with CRIS profile

Christie Alappat Regionales Rechenzentrum Erlangen (RRZE) Jan Laukemann Professur für Höchstleistungsrechnen Thomas Gruber Regionales Rechenzentrum Erlangen (RRZE) Georg Hager Regionales Rechenzentrum Erlangen (RRZE) Gerhard Wellein Professur für Höchstleistungsrechnen

Involved external institutions

Universität Regensburg DE Germany (DE)

How to cite

APA:

Alappat, C., Laukemann, J., Gruber, T., Hager, G., Wellein, G., Meyer, N., & Wettig, T. (2020). Performance Modeling of Streaming Kernels and Sparse Matrix-Vector Multiplication on A64FX. In Proceedings of PMBS 2020: Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (pp. 1-7). Institute of Electrical and Electronics Engineers Inc..

MLA:

Alappat, Christie, et al. "Performance Modeling of Streaming Kernels and Sparse Matrix-Vector Multiplication on A64FX." Proceedings of the 2020 IEEE/ACM Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems, PMBS 2020 Institute of Electrical and Electronics Engineers Inc., 2020. 1-7.

BibTeX: Download