Scheduling of Dependent Tasks in Cloud Computing to VM Tolerant Fault

Cloud computing refers to the use of memory and computing capabilities of computers and servers around the world, where the user has considerable computing power without the need of having powerful machines. The probability of failure occur during the execution becomes stronger when the number of node increases ; since it is impossible to fully prevent failures, one solution is to implement fault tolerance mechanisms. Fault tolerance has become a major task for computer engineers and software developers because the occurrence of faults increases the cost of using resources. In this work we have developed a fault tolerant architecture to Cloud Computing. We have proposed two approaches ; the first approach is essentially based on the checkpoint mechanism in order to minimize the time lost by failures while the second approach is a predection (Deep Learning) which ensures reduce the time lost by faillures in first approach and continuity of service of Cloud Computing in a way efficient in case of failure, The results obtained by the simulation show the effectiveness of our approaches to fault tolerance in term of execution time and masking effects of failures.