%0 Journal Article
%T Checkpointing with Synchronized Clocks in Distributed Systems
%A S. Neogy
%A A. Sinha
%A P. K. Das
%J International Journal of UbiComp
%D 2010
%I Academy & Industry Research Collaboration Center (AIRCC)
%X The processes of the distributed system considered in this paper use loosely synchronized clocks. The paperdescribes a method of taking checkpoints by such processes in a truly distributed manner, that is, in theabsence of a global checkpoint coordinator. The constituent processes take checkpoints according to theirown clocks at predetermined checkpoint instants. A global consistent set of such asynchronous checkpointsneeds to be formed to avoid the domino effect. This is achieved by adding suitable information to theexisting clock synchronization messages looking at which the processes synchronize their checkpoints toform a global consistent checkpoint. Communication in this system is synchronous, so, processes may beblocked for communication at checkpointing instants. The blocked processes save the state they were in justbefore being blocked. It is shown here that the set of such i-th checkpoints is consistent and hence therollback required by the system in case of failure is only up to the last saved state.
%K Fault tolerance
%K Checkpointing
%K Rollback recovery
%K Synchronized clock
%K Clock synchronization message
%U http://www.airccse.org/journal/iju/papers/0410iju5.pdf