摘要
为满足云工作流实例的多样化需求,根据工作流的特点和云环境中资源部署结构,建立多服务质量指标的云工作流调度模型。对蚁群算法进行改进,解决其收敛速度慢、易陷入局部最优等缺点。利用用户对服务质量不同程度的偏好,引入云任务优先次序启发式规则,提出一种基于服务质量的云工作流调度算法(SPACO)。在Cloud Sim平台上,对云工作流调度模型和算法进行仿真分析,将仿真结果与基本蚁群算法(ACO)、改进的蚁群算法(PACO)进行比较,其结果表明该算法能缩短执行时间、降低能耗成本,验证了该模型的可行性和算法的有效性。
To meet various needs of cloud workflow instances,according to the workflow's characteristics and resource deployment structure in cloud environment,a cloud workflow scheduling model with multiple service quality indicators was established.The ant colony algorithm was improved to solve the problems such as the lower convergence rate and easiness to fall into local optimization.Different degrees preferences of users to service quality were used,a cloud workflow task priority heuristic rule was introduced,and a cloud workflow scheduling model based on service quality(SPACO) was proposed.The cloud workflow sche-duling model and algorithm were simulated and analyzed on the CloudSim platform,and the simulation results were compared using ACO and PACO.Comparing results show that the proposed algorithm has shorter execution time and lower cost,which demonstrates the feasibility of the designed model and the effectiveness of the proposed algorithm.
出处
《计算机工程与设计》
北大核心
2018年第1期151-158,259,共9页
Computer Engineering and Design
基金
国家自然科学基金项目(61363001)
宁夏自然科学基金项目(NZ14111)
校级研究生创新基金项目(YCX1658)
关键词
云计算
服务质量
工作流调度
蚁群优化算法
资源优化
cloud computing
quality of service
workflow scheduling
ant colony optimization algorithm
resource optimization
