Abstract
The time warp mechanism is one of the most favorite paradigm to carry out a distributed discrete event simulation, where the synchronization of events can be determined a priori when the program is being executed. The forward computation performed optimistically has to be rolled back when an out-of-order synchronization message is received. This excessive amount of rollback often haunts the performance of a time warp mechanism. We propose in this paper that the semantics of the synchronization messages being sent in a time warp execution can be exploited to reduce the amount of rollback. In particular, we illustrate with some examples how the commutativity and dependency properties of messages can be used to avoid certain types of rollback. For example, two commutative messages can be processed in any order and the resultant computation remains valid. We present a protocol for message handling in the time warp mechanism. We also give a correctness argument for the protocol, leaving the formal proof in the full paper. In the discussion, we suggest several optimization techniques that can be used in our protocol for a better improvement.KeywordsObject StateParallel ProgramDiscrete Event SimulationOperation SequenceState QueueThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.