Abstract

Cloud computing is a promising way providing users computing resources. These resources are provided by means of standard computing instances. Currently there are three price schemas including spot, reservation and on-demand in cloud market, but the average difference among these three pricing schema can be as much as 2.7 times even for the same instance type. In the premise of ensuring service level object, how many and which compute instance are needed, which price schema is needed is important for company. In this paper, an intelligent capacity planning model was proposed. Experimental result shows that our model can save more money while the quality of service does not degrade.