arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1608.00571cs.DCcs.OScs.PL
收藏

TREES:一个具有显式历元同步的CPU/GPU任务并行调度器

TREES: A CPU/GPU Task-Parallel Runtime with Explicit Epoch Synchronization

Blake A. Hechtman, Andrew D. Hilton, Daniel J. Sorin

摘要:

我们已经开发了一个任务并行运行时系统,称为TREES,这是专为CPU/GPU平台上的高性能。在具有多个CPU的平台上,Cilk的“工作优先”原则是任务并行应用程序如何实现性能的基础,但工作优先不适合GPU。我们以工作第一为基础,创建了“协同工作”原则,以解决GPU的具体优势和劣势。协同工作原则扩展了工作优先原则,指出:(a)关键路线上的间接费用应由整个系统立即支付;(b)工作间接费用应由合作支付。我们在OpenCL中实现了TREES运行时,并在CPU/GPU平台上对TREES应用程序进行了实验评估。

英文摘要:

We have developed a task-parallel runtime system, called TREES, that is designed for high performance on CPU/GPU platforms. On platforms with multiple CPUs, Cilk's "work-first" principle underlies how task-parallel applications can achieve performance, but work-first is a poor fit for GPUs. We build upon work-first to create the "work-together" principle that addresses the specific strengths and weaknesses of GPUs. The work-together principle extends work-first by stating that (a) the overhead on the critical path should be paid by the entire system at once and (b) work overheads should be paid co-operatively. We have implemented the TREES runtime in OpenCL, and we experimentally evaluate TREES applications on a CPU/GPU platform.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑