In time warp parallel simulation, the state of each process must be saved (checkpointed) regularlly in case a rollback in necessary. Allghough most existing time warp implementations checkpoint after every state transtition, This is not necessary, and the checkpoint interval is in reality a tuning parameter of the simuation. Lin and Lazowaka proposed a model to divice the optimal checkpoint interval by assuming that the rollback behavior of time warp is not effected by the frequency of checkpointing in general, and that the LIn-Lazowska model my not reflect the real situations in general. This paper extended the lIn-Lazowska model to include the effect of the checkpoint interval on the rollback behavior. The relationship among the overhead associated with state saving and restoration isdescribed. A checkpoint interval selection algorithm wich quickly determines the optimal chcekpoint interval during the execution of time warps simuation is proposed. Empirical results indicate that the algorithms converge quickly and always slsects the optimal checkpoint interval.