Platform Modeling and Scheduling Game with Multiple Intelligent Cloud-Computing Pools for Big Data
We model a generalized game platform with multiple intelligent cloud-computing pools and parallel-queues for rate resources-competing users.Big Data streams to the platform are modeled by triply stochastic renewal reward processes(TSRRPs),where each user may be served simultaneously by multiple pools while each pool with parallel-servers may also serve multi-users at the same time.Our aim in modeling the platform is to model its performance measures(workload and queue length processes)as reflecting diffusion with regime-switchings(RDRSs)under different service policies,e.g.,a Nash equilibrium point myopically at each fixed time to a game problem.Then,by these RDRS models,we can prove our game-based policy to be an asymptotic Pareto minimal-dual-cost Nash equilibrium one globally over the whole time horizon to a randomly evolving dynamic game problem in an asymptotic sense.Iterative schemes for simulating our multi-dimensional RDRS models are also developed with the support of numerical comparisons to illustrate the effectiveness of our RDRS performance models and our game-based policy.
Platform and RDRS modeling Scheduling game Pareto optimal Nash equilibrium Queueing network Big Data Cloud-computing with multiple pools
Wanyang Dai
Department of Mathematics and State Key Laboratory of Novel Software Technology Nanjing University,Nanjing 210093,China
国际会议
河北秦皇岛
英文
181-184
2017-08-21（万方平台首次上网日期，不代表论文的发表时间）