一种基于虚拟截止时间制导的改进的Min-Min元任务调度算法

在网格环境下，资源状况和用户行为相当复杂，是一个异构计算环境，元任务（meta—task）调度比传统并行调度更为复杂。如何映射一组任务到一组机器上被证明是NP问题，其目的一般是最小化任务完成时间（makespan）。为解决这一问题，已经提出一些启发式任务调度算法，例如具有代表性的MinMin元任务调度算法。本文在Min-Min元任务调度算法的基础上，通过虚拟截止时间制导的方法来改进Min-Min算法。实验结果表明，本文提出的算法具有更短的任务完成时间。

A Virtual Deadline Guided Min-Min Meta-task Scheduling Heuristic

In Grid environment which is a Heterogeneous Computing(HC) environment, resource status and user behavior are very complicated, so scheduling heuristics for meta-task are more complicated than traditional parallel scheduling heuristics. How to map a set of tasks on a set of machines is known to be NP-hard. The goal of those scheduling heuristics is minimize the makespan of the meta-tasks. Some heuristics are introduced to solve such scheduling problem, including Min-Min and other heuristics. In this paper, based on traditional Min-Min scheduling heuristic, a virtual deadline guided meta-task scheduling heuristic is proposed. The simulation results show that the proposed heuristic has less makespan than traditional Min-Min scheduling heuristic.