面向云计算数据中心的弹性资源分配算法

Elastic Resource Allocation Algorithm for Cloud Computing Data Center

巨大规模的数据资源与实时多变的应用请求,增加了云计算数据中心的资源分配难度,为此提出一种弹性资源分配算法。通过分析云计算数据中心下弹性资源分配问题,利用带宽资源与中间变量,将分配问题转换为整数线性规划问题,在各链路上逐级划分数据中心后得到不同分区,并按照从下到上的顺序逐层展开运算,完成动态规划阶段,基于树状数据中心的遍历过程,根据极大允许负载占比,明确可行的分配策略,依据带宽需求按序分配,实现云计算数据中心弹性最大化与资源最佳分配。选取不同的基准测试作业作为检测数据,经过对比分析实验数据,验证所提算法在弹性方面具有显著的优越性,有效降低应用请求响应时长,执行效果较为理想。

Massive data resource and time-variant application requests increase the difficulty of resource allocation in cloud computing data center.Therefore,this article put forward an elastic resource allocation algorithm.After analyzing the elastic resource allocation problem in cloud computing data center,we used the bandwidth resources and intermediate variables to transform the allocation problem into an integer linear programming problem.After dividing the data center on each link,we got different regions,and calculated from bottom to top.Then,the dynamic planning stage was completed.Based on the traversal process of tree data center and the maximum allowable load ratio,we determined the feasible allocation strategy,and allocated in order according to the bandwidth demand.Finally,we realized the maximum elasticity and optimal resource allocation of cloud computing data center.Different benchmarking tests were selected as the detection data.The comparative analysis of experimental data verified the significant advantages of the proposed algorithm in flexibility.Meanwhile,this algorithm can effectively reduce the response time of application requests,and its execution effect is ideal.