Cloud scheduling algorithm based on fuzzy clustering
Xuezeng Pan · Journal of Communications · 2012
In order to better serve cloud's objective of providing low-cost and on-demand services,a two-level and fuzzy clustering based cloud scheduling algorithm named FCTLBS(fuzzy clustering and two level based task scheduling algo-rithm) was proposed.Scheduling was divided into two levels: user scheduling and task scheduling.It used fuzzy cluster-ing method to cluster resources according to their capabilities and computed tasks' resource bias coefficient according to their parameters.Thus tasks with different resource preference were able to choose different capability resources in the corresponding clusters.The new algorithm reduces the scale of resource choice,better reflects cloud tasks' service re-quirements.Simulation results show that the proposed algorithm has some advantages compared to other kindred algo-rithms.