A Ranking Chaos Algorithm for dual scheduling of cloud service and computing resource in private cloud

Abstract

Private cloud as an important branch of cloud computing has brought significant benefit to many kinds of conglomerates in resource sharing. With central management of centre console, Service Composition Optimal Selection (SCOS) and Optimal Allocation of Computing Resources (OACR) are two critical steps for implementing high flexible and agile service provision and resource sharing among sub-enterprises and partner-enterprises under the key technologies of virtualization. However, two steps decision-making are inefficient and cumbersome. To overcome this deficiency, the idea of combining SCOS and OACR into one-time decision in one console is first presented in this paper, named Dual Scheduling of Cloud Services and Computing Resources (DS-CSCR). The mutual relations between the upper layer cloud services and the underlying infrastructures and their properties in the private cloud of conglomerate are deeply analyzed. For addressing large-scale DS-CSCR problem, a new Ranking Chaos Optimization (RCO) is proposed. With the consideration of large-scale irregular solution spaces, new adaptive chaos operator is designed to traverse wider spaces within a short time. Besides, dynamic heuristic and ranking selection are introduced to control the chaos evolution in the proposed algorithm. Theoretical analysis and simulations demonstrate that the new DS-CSCR outperforms the traditional two-level decision making with the improvements in both cloud service composition and computing resource allocation. In addition, RCO can remarkably give much prominent solutions with low time-consuming and high stability than a few typical intelligent algorithms for solving DS-CSCR in private cloud. With the new DS-CSCR and RCO, cloud services and computing infrastructures can then be quickly combined and shared with high efficient decision.