DMCA
Inter-Block GPU Communication via Fast Barrier Synchronization
| Citations: | 39 - 2 self |
BibTeX
@MISC{Xiao_inter-blockgpu,
author = {Shucai Xiao and Wu-chun Feng},
title = {Inter-Block GPU Communication via Fast Barrier Synchronization},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract—While GPGPU stands for general-purpose computation on graphics processing units, the lack of explicit support for inter-block communication on the GPU arguably hampers its broader adoption as a general-purpose computing device. Interblock communication on the GPU occurs via global memory and then requires barrier synchronization across the blocks, i.e., inter-block GPU communication via barrier synchronization. Currently, such synchronization is only available via the CPU, which in turn, can incur significant overhead. We propose two approaches for inter-block GPU communication via barrier synchronization: GPU lock-based synchronization and GPU lock-free synchronization. We then evaluate the efficacy of each approach via a micro-benchmark as well as three well-known algorithms — Fast Fourier Transform (FFT), dynamic programming, and bitonic sort. For the microbenchmark, the experimental results show that our GPU lockfree synchronization performs 8.4 times faster than CPU explicit synchronization and 4.0 times faster than CPU implicit synchronization. When integrated with the FFT, dynamic programming, and bitonic sort algorithms, our GPU lock-free synchronization further improves performance by 10%, 26%, and 40%, respectively, and ultimately delivers an overall speed-up of 70x, 13x, and 24x, respectively. I.
Keyphrases
inter-block gpu communication fast barrier synchronization gpu lock-free synchronization dynamic programming improves performance gpu lock-based synchronization bitonic sort algorithm bitonic sort general-purpose computation explicit support general-purpose computing device cpu explicit synchronization interblock communication overall speed-up well-known algorithm fast fourier transform global memory gpu lockfree synchronization experimental result cpu implicit synchronization inter-block communication
