Communication-Aware Scheduling of Precedence-Constrained Tasks

Communication-Aware Scheduling of Precedence-Constrained Tasks
复制标题

优先级受限任务的通信感知调度

DOI:
--
复制
发表时间:
2019
期刊:
Sigmetrics Performance Evaluation Review
影响因子:
--
通讯作者:
Yuxiong He
Yuxiong He
中科院分区:
--
文献类型:
--
作者:
Yu Su;Xiaoqi Ren;Shai Vardi;A. Wierman;Yuxiong He

文献摘要

被引文献

相似文献

大规模机器学习平台中的作业使用具有优先约束的任务计算图来表示。为了在异构服务速率和通信时间设置中处理具有机器依赖通信需求的优先级约束任务,我们提出了一种新的调度框架,即广义最早时间优先(GETF),该框架改进了该领域的最新结果。具体来说,我们提供了第一个可证明的最坏情况近似保证,以最小化具有优先约束的任务在具有机器依赖的通信时间的相关机器上的最大完工时间。
Jobs in large-scale machine learning platforms are expressed using a computational graph of tasks with precedence constraints. To handle such precedence-constrained tasks that have machine-dependent communication demands in settings with heterogeneous service rates and communication times, we propose a new scheduling framework, Generalized Earliest Time First (GETF), that improves upon stateof- the-art results in the area. Specifically, we provide the first provable, worst-case approximation guarantee for the goal of minimizing the makespan of tasks with precedence constraints on related machines with machine-dependent communication times.