Weighted Service Broker Algorithm in Cloud Environment

Cloud Computing refers to on-demand delivery of IT resources and applications via the Internet with pay-as-you-go pricing. The cloud service provider provides these services through datacenters with high configured servers. Therefore, many factors are studied such as response time, cost, the number of requests, and choosing the best datacenter (DC) to serve the customer request. Cloud service broker acts as a mediator between customer and service provider and it is responsible for choosing the datacenter. There are many policies to determine this selection such as Service Proximity-Based Routing, which selects the earliest region, which has minimum communication delay and lowest network latency. However, this policy adopts the random selection of datacenter when the closest region includes more than one datacenter; it does not consider any factor such as the cost of the datacenter, and its processing time. Therefore, a weighted based approach aware service brokering policy is proposed to avoid the random selection of datacenters by grading datacenters according to their virtual machine (VM) cost, data transfer cost, and the number of VMs included. The proposed algorithm has been tested and evaluated using Cloud Analyst software, and the results show a significant reduction in the total cost, processing time, and response time.