The Technology Research Of Resource Scheduling Based On OpenStack

The development of cloud computing technology has changed the traditional IT architecture,which has brought the revolutionary change to the management and the utilization of the traditional physical resources.With the widespread use of cloud computing,how to make the resources be fully balanced has become one of the hot issues in resource scheduling research.Improving the virtual machine resource scheduling strategy to solve the problem of load imbalance and the strategy can also improve resource utilization and reduce energy consumption.This thesis mainly studies the virtual machine resource scheduling strategy based on OpenStack.Firstly,this thesis studies the virtual machine scheduling strategy based on network bandwidth.The virtual machine scheduling mechanism of OpenStack is based on cpu,memory and storage resources,without considering the reasonable scheduling of network bandwidth resources.Therefore,this thesis takes into account the constraints of network bandwidth,and extends the virtual machine scheduling strategy of OpenStack.The simulation results show that the scheduling strategy can improve the network bandwidth of virtual machine more effectively than default scheduling strategy in OpenStack.Secondly,a virtual machine scheduling strategy based on multi-objective fusion algorithm of particle swarm and ant colony is proposed to solve the multi-objective constraint problem in OpenStack,which can find a compromise among multiple contradictory goals.Multi-objective fusion algorithm is an improvement of particle swarm and ant colony algorithm.This thesis proposes a model of resource waste,energy loss and load imbalance for physical clusters and the Parato optimal solution of virtual machines mapped to the physical host is accelerated by algorithm fusion.The simulation results show that the proposed multi-objective fusion algorithm can improve the multidimensional resource utilization,reduce the energy consumption and load imbalance compared with the traditional heuristic algorithm particle swarm and ant colony algorithm.Finally,this thesis designs and implements the virtual machine scheduling system platform based on OpenStack.The monitoring module and the database module are mainly designed to monitor the resources utilization.Optimizing the utilization and load balancing of multidimensional resources by optimizing the virtual machine scheduling algorithm for the next deployment location.