USE OF COMMON TIME BASE FOR CHECKPOINTING AND ROLLBACK RECOVERY IN A DISTRIBUTED SYSTEM

A new approach for checkpointing and rollback recovery in a distribute d computing system using a common time base is proposed in this paper. First, a common time base is established in the system using a hardwa re clock synchronization algorithm. This common time base is coupled w ith the idea of pseudo-recovery points to develop a checkpointing algo rithm that has the following advantages: 1) reduced wait for commitmen t for establishing recovery lines, 2) fewer messages to be exchanged, and 3) less memory requirement. These advantages are assessed quantita tively by developing a probabilistic model.