Dynamic Resizing on Active Warps Scheduler to Hide Operation Stalls on GPUs

Myung Kuk Yoon, Yunho Oh, Seung Hun Kim, Sangpil Lee, Deokho Kim, Won Woo Ro

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

This paper conducts a detailed study of the factors affecting the operation stalls in terms of the fetch group size on the warp scheduler of GPUs. Throughout this paper, we reveal that the size of a fetch group is highly involved for hiding various types of operation stalls: Short latency stalls, long latency stalls, and Load/Store Unit (LSU) stalls. The scheduler with a small fetch group cannot hide short latency stalls due to the limited number of warps in a fetch group. In contrast, the scheduler with a large fetch group cannot hide long latency and LSU stalls due to the limited number of fetch groups and the lack of memory subsystems, respectively. To hide various types of stalls, this paper proposes a Dynamic Resizing on Active Warps (DRAW) scheduler which adjusts the size of a fetch group dynamically based on the execution phases of applications. For the applications that have the best performance at LRR (one fetch group), the DRAW scheduler matches the performance of LRR and outperforms TL (multiple fetch groups) by 22.7 percent. In addition, for the applications that have the best performance at TL, our scheduler achieves 11.0 and 5.5 percent better performance compared to LRR and TL, respectively.

Original languageEnglish
Article number7927466
Pages (from-to)3142-3156
Number of pages15
JournalIEEE Transactions on Parallel and Distributed Systems
Volume28
Issue number11
DOIs
StatePublished - 1 Nov 2017

Keywords

  • General Purpose on GPUs (GPGPUs)
  • Graphics Processing Units (GPUs)
  • warp scheduler

Fingerprint

Dive into the research topics of 'Dynamic Resizing on Active Warps Scheduler to Hide Operation Stalls on GPUs'. Together they form a unique fingerprint.

Cite this