MGPUSim: Enabling Multi-GPU Performance Modeling and Optimization
Northeastern University
International Symposium on Computer Architecture (ISCA), 2019
@article{mccardwell2019mgpusim,
title={MGPUSim: Enabling Multi-GPU Performance Modeling and Optimization},
author={McCardwell, Yuhui Bao1 Spencer Hance1 Carter and Barclay, Vincent Zhao1 Harrison and Rafael, Amir Kavyan Ziabari3 Zhongliang Chen and Kim, Ubal1 Jos{‘e} L Abell{‘a}n4 John and Joshi, Ajay and Kaeli, David},
year={2019}
}
The rapidly growing popularity and scale of data-parallel workloads demand a corresponding increase in raw computational power of Graphics Processing Units (GPUs). As single-GPU platforms struggle to satisfy these performance demands, multi-GPU platforms have started to dominate the high-performance computing world. The advent of such systems raises a number of design challenges, including the GPU microarchitecture, multi-GPU interconnect fabric, runtime libraries, and associated programming models. The research community currently lacks a publicly available and comprehensive multi-GPU simulation framework to evaluate nextgeneration multi-GPU system designs. In this work, we present MGPUSim, a cycle-accurate, extensively validated, multi-GPU simulator, based on AMD’s Graphics Core Next 3 (GCN3) instruction set architecture. MGPUSim comes with in-built support for multi-threaded execution to enable fast, parallelized, and accurate simulation. In terms of performance accuracy, MGPUSim differs by only 5.5% on average from the actual GPU hardware. We also achieve a 3.5x and a 2.5x average speedup running functional emulation and detailed timing simulation, respectively, on a 4-core CPU, while delivering the same accuracy as serial simulation. We illustrate the "exibility and capability of the simulator through two concrete design studies. In the first, we propose the Locality API, an API extension that allows the GPU programmer to both avoid the complexity of multi-GPU programming, while precisely controlling data placement in the multi-GPU memory. In the second design study, we propose Progressive Page Splitting Migration (PASI), a customized multi-GPU memory management system enabling the hardware to progressively improve data placement. For a discrete 4-GPU system, we observe that the Locality API can speed up the system by 1.6x (geometric mean), and PASI can improve the system performance by 2.6x (geometric mean) across all benchmarks, compared to a unified 4-GPU platform.
June 23, 2019 by hgpu
