基础设施服务层资源分配的通用逐步优化策略

Universal and Gradual-optimization Strategies for Resource Allocation in Infrastructure as a Service

为了提高各类资源利用率,在分析基础设施服务层计算、存储、网络资源共性的基础上,提出以高效解决0-1背包问题为靶向的通用逐步优化策略;设计时间复杂度分别为多项式级别的基于贪婪算法、进化算法、线性规划算法的通用逐步优化策略;通过理论分析及实验验证,得出基于3类优化算法的通用逐步优化策略的优缺点及适用范围。结果表明:贪婪算法能够满足时间方面的要求,进化算法能够满足时间、单个资源利用率和整体资源利用率3个方面的要求,线性规划算法能够满足单个资源利用率和整体资源利用率2个方面的要求。

To improve the utilization ratio of various resources,a universal and gradual-optimization strategy taking efficiently solving the 0-1 knapsack problem as targeting was proposed based on common characteristics of computing resources,storage resources,and network resources in infrastructure as a service.The universal and gradual-optimization strategy based on greedy algorithm,evolutionary algorithm,and linear programming algorithm were designed respectively with the time complexity of polynomial level.By theoretical analysis and the experimental verification,the merits and demerits as well as scope of application of the universal and gradual-optimization strategy based on three algorithms were obtained.The results show that the greedy algorithm can meet the requirements of time,the evolutionary algorithm can meet the requirements of time,single resource utilization and overall resource utilization,and the linear programming can satisfy both the requirements of single resource utilization and overall resource utilization.