The Dynamic Scheduling And Management Of Computing Resource In Cloud Platform Based On OpenStack

Cloud computing is a new generation IT model, evolved from grid computing, parallel computing, and distributed computing. Users can use it easily via network to access to a configurable computing resources on-demand (e.g. networks, servers, storage, shared pool of applications and services, etc.) by the minimal management effort or service provider intervention. So that it can be quickly opened and release resources. Now the cloud environment resources are made up through virtualization technology underlying hardware resources virtualization, forming a huge virtual resource pool available to the user. With the use of cloud computing users continued increase with the size of the cloud data center also continued to increase, and how to make efficient use of cloud virtualization resources and quickly available to the user, reducing users’waiting time and improving the entire cloud data utilization of the center, which has become a cloud computing environment in a virtual machine resources dynamic scheduling problems.This paper studies the cloud data center virtualization, dynamic resource scheduling policy, based on the most popular open source cloud to computing platform OpenStack expand the basis of previous work, following a series of work and innovations:(1) Analysis of the current the basic characteristics of cloud computing, the architecture and the key technology, comparing with several open source cloud computing platform, and carried out a detailed study of the cloud data center resource scheduling and management needs and key technologies;(2) Based on OpenStack architecture of virtualized resources modeling, described from the level of service and resource levels of resources, and proposed dynamic real-time feedback integrated load balancing scheduling strategies and algorithms through the CPU, memory, storage and network bandwidth load four dimensions.As well as calculating the cloud platform inequality of data centers and the physical servers. By comparing with the round-robin scheduling algorithm, and the minimum link algorithm in cloudsim simulation platform, the experimental results show that the proposed algorithm enables the application of virtual machine instances for better deployment location, but also to the cloud data center could achieve a more load balancing resources to prove the effectiveness and stability of the algorithm; (3) Combining with cluster management tools xCAT and making a secondary development, the resource automated management platform is implemented under the OpenStack environment, being able to effectively monitor the management of the resources, the dynamic expansion of the resources in the cloud environment in an automated way, to achieve large-scale automated operation and bare-metal provisioning management.