Data Center Workload Management of Scheduling, Resource Processing

Abstract

To offering a pool of computation, communication and storage resources are undoubtedly the most efficient platforms to meet the ever-growing needs of BDSP, Different datacenter pairs are with different inter-datacenter network costs charged by Internet Service Providers (ISPs). Although, inter-datacenter transportation in BDSP establishes a large amount of a cloud provider’s traffic demand over the Internet and suffers significant communicative cost, which may even become the dominant operational expenditure factor. As the datacenter resources are provided in a virtualized way, the virtual machines for stream processing tasks can be freely deployed onto any datacenters, provided that the Service Level Agreement i.e. quality-of-information is obeyed. In this paper we propose a general modeling framework that describes all representative inter-task relationship semantics in BDSP. Based on our novel framework, later on we provide a traffic minimization problem for BDSP into a mixed-integer linear programming and prove it to be NP-hard. Then it offers a computation-efficient solution based on MILP. The extensive simulation will validate the efficiency of this proposal.