Request generation and intelligent scheduling for cloud educational resource datacenter

This paper presents the architecture, the simulation model and scheduling algorithms for a cloud educational resource datacenter (CERD). CERD is a way to provide an economically profitable shared remote access to paid software for educational facilities. The architecture reflects the physical and logical structure of CERD and its components. The simulation model describes the simulation scheme and the method for generation and processing of user requests. The problem of efficient CERD scheduling for the optimal use of cloud virtual machines and software licenses have been studied in details. Within the research two intelligent algorithms for scheduling are implemented: the simulated annealing algorithm and genetic algorithm. The algorithm estimation criteria are the performance time and the count of satisfied requests. The most efficient algorithm will be used in further experimental researches on actual CERD implementation.