MaxPair: Enhance OpenCL Concurrent Kernel Execution by Weighted Maximum Matching

Yuan Wen, Michael F. P. O'Boyle, Christian Fensch

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)
30 Downloads (Pure)

Abstract

Executing multiple OpenCL kernels on the same GPU concurrently is a promising method for improving hardware utilisation and system performance. Schemes of scheduling impact the resulting performance significantly by selecting different kernels to run together on the same GPU. Existing approaches use either execution time or relative speedup of kernels as a guide to group and map them to the device. However, these simple methods work on the cost of providing suboptimal performance.

In this paper, we propose a graph-based algorithm to schedule co-run kernel in pairs to optimise the system performance. Target workloads are represented by a graph, in which vertices stand for distinct kernels while edges between two vertices represent the corresponding two kernels co-execution can deliver a better performance than run them one after another. Edges are weighted to provide information of performance gain from co-execution. Our algorithm works in the way of finding out the maximum weighted matching of the graph. By maximising the accumulated weights, our algorithm improves performance significantly comparing to other approaches.
Original languageEnglish
Title of host publicationProceedings of the 11th Workshop on General Purpose Processing Using GPUs
PublisherAssociation for Computing Machinery
Pages40-49
Number of pages10
ISBN (Electronic)9781450356473
DOIs
Publication statusPublished - 24 Feb 2018
Event11th Workshop on General Purpose Processing Using GPUs - Vienna, Austria
Duration: 25 Feb 201825 Feb 2018

Conference

Conference11th Workshop on General Purpose Processing Using GPUs
CountryAustria
CityVienna
Period25/02/1825/02/18

Keywords

  • GPGPU
  • Scheduling
  • Concurrent Kernels

Fingerprint Dive into the research topics of 'MaxPair: Enhance OpenCL Concurrent Kernel Execution by Weighted Maximum Matching'. Together they form a unique fingerprint.

  • Cite this

    Wen, Y., O'Boyle, M. F. P., & Fensch, C. (2018). MaxPair: Enhance OpenCL Concurrent Kernel Execution by Weighted Maximum Matching. In Proceedings of the 11th Workshop on General Purpose Processing Using GPUs (pp. 40-49). Association for Computing Machinery. https://doi.org/10.1145/3180270.3180272