并行Open-Shop和优先约束调度模型的流程时间优化
Flow-time Optimization For Concurrent Open-Shop and Precedence Constrained Scheduling Models
摘要：
在一组机器上调度一组作业是一个基本问题，每天需要在各种计算平台上解决数百万次：在操作系统中，在大型数据集群中，以及在数据中心中。与最大完工时间一样，流时间（flow-time）测量了作业在完成之前在系统中花费的时间长度，可以说是衡量调度算法性能的最重要指标。近年来，在理解基于流时间的目标函数在不同的设置，如不相关的机器调度，广播调度，多维调度，仅举几例，已经取得了显着的进展。然而，我们对流程时间目标的理解主要限于作业具有简单结构的场景;特别是，每个作业都是一个独立的实体。另一方面，在几乎所有现实世界的应用程序中，例如MapReduce设置，作业具有更复杂的结构。在本文中，我们考虑了两个经典的调度模型，捕捉复杂的作业结构：1）并发开放车间调度和2）优先约束调度。我们研究这些问题的主要动机具体来自于它们与两个调度问题的相关性，这两个调度问题在数据中心的背景下已经变得非常重要：同流调度和DAG调度。我们设计了几乎最优的近似算法来求解开放式车间调度和优先约束调度问题，并给出了相应的结果。
英文摘要：
Scheduling a set of jobs over a collection of machines is a fundamental problem that needs to be solved millions of times a day in various computing platforms: in operating systems, in large data clusters, and in data centers. Along with makespan, flow-time, which measures the length of time a job spends in a system before it completes, is arguably the most important metric to measure the performance of a scheduling algorithm. In recent years, there has been a remarkable progress in understanding flow-time based objective functions in diverse settings such as unrelated machines scheduling, broadcast scheduling, multi-dimensional scheduling, to name a few. Yet, our understanding of the flow-time objective is limited mostly to the scenarios where jobs have simple structures; in particular, each job is a single self contained entity. On the other hand, in almost all real world applications, think of MapReduce settings for example, jobs have more complex structures. In this paper, we consider two classical scheduling models that capture complex job structures: 1) concurrent open-shop scheduling and 2) precedence constrained scheduling. Our main motivation to study these problems specifically comes from their relevance to two scheduling problems that have gained importance in the context of data centers: co-flow scheduling and DAG scheduling. We design almost optimal approximation algorithms for open-shop scheduling and precedence constrained scheduling, and show hardness results.
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。
