Cloud Platform Resources Scheduling Problem Research Based On Internet Of Vehicles Application

Cloud computing platform based on IOV(Internet of Vehicles). application integrates different types of heterogeneous resources, such as physical servers and virtual machines, into a single virtual resource pools, using virtualization technology, providing different needs with different types of IOV application service. Most specific IOV application service has many characteristics, such as strong real-time performance, and these characteristics has higher requirements on the IOV cloud computing platform. Resource scheduling is one of the important hot topics in the field of cloud computing, which will directly affects the stability and service reliability of cloud platform. Therefore, the research on resource-scheduling algorithm for IOV cloud platform can be very significant in the theory and reality.This paper studies the resource scheduling of cloud computing problem under the background of IOV application from the two following aspects.(1) The application of IOV have the characteristics of multi-user, multi-service, high concurrency and so on. In order to ensure the needs of the rapid, stable and reliable operation of IOV on the cloud platform, a task scheduling algorithm MCT-LB-GSA (Minimum Completion Time-Load Balance-Greedy Scheduling Algorithm) is proposed based on the application of IOV. This algorithm puts the task scheduling to the virtual machine resources which has lighter current load and minimum completion time according to the greedy strategy. The Simulation results in the CloudSim show that the MCT-LB-GSA algorithm ensures the optimal task scheduling span, at the same time realizes the resource load balancing and improves the resources utilization effectively.(2) Large amount of traffic data of IOV need a powerful platform to process and analyze, Cloud data center just meets the needs of IOV. However, if the applicative characteristic of IOV is inconsistent with the Cloud data center physical computer configuration, the load imbalance comes. In order to solve the problem, the following contents raise the CDM-CU(Combining the distance matching and comprehensive utilization) virtual machine deployment algorithm on the base of IOV. The algorithm neither seeks the best distance between the virtual machine and the physical server, nor the minimum load of data center, instead mix the above two with harmonic factor, and then choose a suitable physical machine for the submit business of customer to deploy corresponding virtual machine set. To do a simulation test in the environment of CloudSim, the results show that this algorithm can make a good loading balance on the base of meeting personalized service.