摘要
提出了一种基于动态重用的成本感知的云服务请求调度算法,不仅能够根据当前的系统负载状况和云服务请求的可分性特征,按需租用和重用虚拟资源,实现云服务动态请求的优化调度,而且还能在满足服务等级协议约束的同时,最小化虚拟资源租赁成本,增加云服务供应商的利润.仿真实验表明,提出的算法的资源利用率、运营利润均高于用以对比的调度算法.
With help of a new cost - aware cloud service request scheduling algorithm based on dynamic reuse, according to current system load and divisible character of cloud service requests, the virtual re- sources can be rent and reused on demand to achieve optimal scheduling of dynamic requests in reasona- ble time. The rental cost of the overall infrastructure for increasing cloud service providers' profits can be minimized when meeting service level agreement constraints. Simulation indicates that our proposed algo- rithm shows better performance compared with other revenue - aware algorithms in terms of resource utili- zation and operation profit.
出处
《北京邮电大学学报》
EI
CAS
CSCD
北大核心
2013年第1期86-90,共5页
Journal of Beijing University of Posts and Telecommunications
基金
教育部博士点基金项目(20110005130001)
国家自然科学基金项目(61272521)
新世纪优秀人才支持计划项目(NECT-100263)
关键词
云服务
服务等级协议
成本感知
虚拟机
遗传算法
cloud service
service level agreement
cost-aware
virtual machine
generic algorithm