On demand check pointing for grid application reliability using communicating process model

The objective of the work is to propose an on-demand asynchronous check pointing technique for the fault recovery of a grid application in communicating process approach. The formal modelling of processes using LOTOS is done wherein the process features are declared in terms of possibilities of rollback and replicas permitted to accept the assigned tasks as decided by the scheduler. If any process is tending to be faulty in run time that will be detected by check pointing mechanism through the Task Dependency Graph (TDG) and their respective worst case execution time and dead line parameters are used to decide the schedulability. The Asynchronous Check Pointing On Demand (ACP-OD) approach is used to enhance the grid application reliability through the needed fault tolerant services. The scheduling of concurrent tasks can be done using the proposed Concurrent Task Scheduling Algorithm (CTSA) algorithm to recover from the faulty states using replication or rollback techniques. The check pointing and replication mechanisms have been used in which the synchronization between communicating processes is needed to enhance the efficiency of check pointing mechanism. The model is tested with a number of rollback variables treating the application as a Stochastic Activity Network (SAN) using Mobius.