云环境下结合改进粒子群优化与检查点技术的容错调度算法  

Fault-tolerant Scheduling Algorithm Based on Improved Particle Swarm Optimization in Cloud Environment

作　　者：孙默辰 邵清[1] SUN Mo-chen;SHAO Qing(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学光电信息与计算机科学学院,上海200093

出　　处：《软件导刊》2020年第2期7-11,共5页Software Guide

基　　金：国家自然科学基金项目(61703278);上海市科委科研计划项目(17511107203)。

摘　　要：云计算环境中任务执行容易受资源故障影响,导致调度效率与成功率降低。针对该问题,提出一种结合改进粒子群优化与检查点技术的容错调度算法。通过改进粒子群优化算法进行全局搜索,寻找粒子群最优解,以保证任务获取最优资源,减少调度复杂度;同时通过设置检查点,使失效任务从检查点继续执行,实现任务动态恢复,提高调度可靠性。仿真实验表明,与传统算法相比,当任务数量不断增加时该算法可提高任务执行成功率,缩短任务执行时间。Task execution is susceptible to resource failures in cloud computing environment which leads to a reduction in scheduling efficiency and success rate.In order to solve this problem,a fault-tolerant scheduling algorithm based on improved particle swarm optimization and checkpoint technology is proposed.Firstly,the particle swarm optimization algorithm is used for global search to find the optimal solution,and it will ensure that the task gets the optimal resources and reduces the complexity of the task scheduling.At the same time,by setting a checkpoint,the invalid task is continuously executed from the checkpoint to achieve dynamic recovery of the task and improve scheduling reliability.Simulation experiments show that as the number of tasks increases,the algorithm can improve the success rate of task execution and also shorten the task execution time when compared with the traditional algorithm.

关 键 词：云计算 任务调度 容错 粒子群优化 检查点技术 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]