Automated Deployment System In Cloud Data Center

With the development of cloud computing, the scale of data center in cloudenvironment enlarges gradually, which brings more computational workload to thedeployment. The deployment of data center includes the hardware resources (server),software resources (middleware or database), network resources and storage resources.Currently, cloud data center mainly focused on the speed of deployment, as well as afriendly graphic user interface. But it ignores the reliability and availability of deploymentserver, which will affect the system’s on-line business and result in unnecessary losses. So,how to deploy the resources automatically in cloud data center as well as guarantee theservice usability of deployment has become a hot topic in current research work.In response to these issues, data center automatically deployment systems is proposedto face cloud computing environment. Then we present automatically deployment systemin cloud data center with the characteristics of reliability, availability and load-balance.Petri workflow engine can deal with the deployment process dependencies effectively. Weuse RAID1, Bonding, Heartbeat and kickstart in order to avoid single point failure ofdeployment server and propose two kind of load balancing algorithm to solve thebottleneck problem of deploying the server in large-scale cloud environment.The experimental results show that Petri workflow engine complete automaticdeployment of physical resources in cloud data center. The performance experiment showthat the nodes deployment time are significantly lower than the average system installationand the success rate of deployment can keep on90%in a high concurrency.