A solution based on prediction for cloud resources allocation

Abstract

The phenomenon of the clouds makes it possible to have a dynamic resource management. The effectiveness of this management becomes an important issue with the increasing size of the number of resources and the number of users. Because of a non-negligible initialization cost, a precise prediction needs must be made. In this article, we offer an approach to the prediction of the Identification based on the use of resources in the history of a platform. We represent the algorithm of resources allocation witch uses this method to assist the system to use resource allocation decisions as well as results of utilization expertise of cloud resources historical load.