KAWS: Coordinate Kernel-Aware Warp Schedulingand Warp Sharing Mechanism for Advanced GPUs


Viet Tan Vo, Cheol Hong Kim, Journal of Information Processing Systems Vol. 17, No. 6, pp. 1157-1167, Dec. 2021  

10.3745/JIPS.01.0084
Keywords: GPU, Multiple Warp Schedulers, Resource Underutilization, Warp Scheduling
Fulltext:

Abstract

Modern graphics processor unit (GPU) architectures offer significant hardware resource enhancements for parallel computing. However, without software optimization, GPUs continuously exhibit hardware resource underutilization. In this paper, we indicate the need to alter different warp scheduler schemes during different kernel execution periods to improve resource utilization. Existing warp schedulers cannot be aware of the kernel progress to provide an effective scheduling policy. In addition, we identified the potential for improving resource utilization for multiple-warp-scheduler GPUs by sharing stalling warps with selected warp schedulers. To address the efficiency issue of the present GPU, we coordinated the kernel-aware warp scheduler and warp sharing mechanism (KAWS). The proposed warp scheduler acknowledges the execution progress of the running kernel to adapt to a more effective scheduling policy when the kernel progress attains a point of resource underutilization. Meanwhile, the warp-sharing mechanism distributes stalling warps to different warp schedulers wherein the execution pipeline unit is ready. Our design achieves performance that is on an average higher than that of the traditional warp scheduler by 7.97% and employs marginal additional hardware overhead.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from November 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.




Cite this article
[APA Style]
Vo, V. & Kim, C. (2021). KAWS: Coordinate Kernel-Aware Warp Schedulingand Warp Sharing Mechanism for Advanced GPUs. Journal of Information Processing Systems, 17(6), 1157-1167. DOI: 10.3745/JIPS.01.0084.

[IEEE Style]
V. T. Vo and C. H. Kim, "KAWS: Coordinate Kernel-Aware Warp Schedulingand Warp Sharing Mechanism for Advanced GPUs," Journal of Information Processing Systems, vol. 17, no. 6, pp. 1157-1167, 2021. DOI: 10.3745/JIPS.01.0084.

[ACM Style]
Viet Tan Vo and Cheol Hong Kim. 2021. KAWS: Coordinate Kernel-Aware Warp Schedulingand Warp Sharing Mechanism for Advanced GPUs. Journal of Information Processing Systems, 17, 6, (2021), 1157-1167. DOI: 10.3745/JIPS.01.0084.