Design principles of request scheduler in cloud

The principle of assigning users' requests with computing resources in large scale distributed computing environments highly impacts the system's performance. Therefore, adopting a suitable requests assignment policy will ascertain to achieve high performance. In this paper, we present generic principles of a request assignment scheduler and its structural components. The proposed principle is based on a load balancing principle that effectively handles a problem scenario of assigning `n' servers serving `m' requests with varying service time, limited by node's multi-constraints. Conceptually, this model can be adopted to any networked service delivery architecture with the careful adoption of parameters in the interested domain. We presented an algorithm for the proposed principle and developed a simulator namely Requests Assignment Scheduler (RAS). This paper also investigates impacting parameters of requests assignment in cloud architectures.