Cloud Computing (CC) represents an attractive and cost-efficient of server-based computing and application service provider models. Virtualization technology enables dynamically allocate resources based on workload fluctuations from users’ needs. The current paradigm of Data Center (DC) or Mega Data Center (MegaDC) has changed to new approach where Micro Data Centers (MDC) are distributed to different locations, Cloud services and solutions are deployed onto those MDCs in order to provide services to enterprise, home or mobile users.
Therefore, by identifying future resources needed before scaling up or down the MDC’s resources pool is one of the most challenge research. In this paper, we propose the resources prediction technique based on historical data at MDC level for allocating the appropriate resources. The experiment results show that by using this technique we can predict the future resource needed for efficient resource provisioning in Cloud Computing environment.