Efficient Resource Scheduling in Data Centers using MRIS

Scheduling the resources is decisive in shared data centers. Today scheduling algorithms focus only on onedimensional resource models, infact the multiple resources (e.g. Memory, Storage, CPU and Network bandwidth) can be consumed concurrently. There is a lot of demand for multiple resources in cloud computing which allows various users to share a data center. This paper depicts resource scheduling problem to a bounded multi-dimensional knapsack problem, taking into account the requirement dependency among multi-dimensional resources. Due to the NP hardness of B-MDKP, we present Multi-dimensional Resource Integrated Scheduling (MRIS), an inquisitive algorithm to obtain the approximate optimal solution. MRIS will be highly efficient and achieves high performance for a various set of workloads.