基于负载分时分析的虚拟服务整合建模分析

Modeling Analysis of Virtual Service Consolidation Based on Workload Timesharing Analyzing

介绍了一种面向下一代绿色数据中心的虚拟服务整合方案。基于服务负载高峰处于不同时段的原理,提出了基于负载分时分析的建模方法,在保证QoS的前提下能够达到资源利用率最大化。考虑了服务之间的联系性和互斥性,以及服务与服务器之间的兼容性,提出了5项整合原则,使该方案具有更强的实际应用价值。将建立的模型看作是有约束的多维装箱问题,提出了基于分组遗传算法(GGA)的智能优化算法搜索全局最优解。通过实验表明,该方案与之前的模型相比具有更高的整合率。

A solution of virtual service consolidation oriented to next generation green data center is introduced in this paper. Due to the peak-valley of the workload of services is in different time, a way to problem modeling is proposed based on workload timesharing analysis, which can achieve maximal resource utilization on the premise of keeping the required QoS. Meanwhile, the associated services and the mutual services, and the compatibility between services and servers are taken into account and five principles of consolidation are considered. By regarding the modeling as a multi-dimension bin packing problem with constraints, we propose a GGA-based heuristic to search for the global optimal solution. The experimental results show that the solution can get a better consolidation ratio than the former model.