Online provisioning for virtual clusters in multi-tenant cloud data center network

Abstract

The rapid development of cloud computing and high demands from network operators require strong support from the underlying data centers. Therefore, the effectiveness of using physical resources has become a target for operators and researchers. This paper discusses the online virtual cluster provision problem with multi-tenant in DCNs, which aims to determine when and where the virtual clusters are placed in the DCN. To address this problem, we formulate it into two parts: online multi-tenant scheduling and virtual cluster placement. Online multi-tenant scheduling aims to determine the execution order for the multi-tenant, and the virtual cluster placement determines the position where virtual machines (VMs) are placed. Our objective is to maximize the revenue for the DCN under the resource constraints. Due to the problem complexity, an efficient heuristic algorithm OMS (Online Resource Scheduling) is proposed to solve the online multi-tenant scheduling problem. We further explore the virtual cluster placement problem based on the OMS, and propose a novel algorithm VMP-VC. We evaluate our proposals through a series of simulations, and the simulation results demonstrate that OMS and VMP-VC can greatly increase the total revenue for the DCN.