Dynamic Performance Profiles Of Simulation Calendars

Abstract

W e propose a suite of tests based on two-state Markov chains f o r experimentally assessing the dynamic performance of a variety of simulation calendar implementations. In contrast t o previous studies based on the standard hold model for evaluation of performance statically, the proposed Markov hold model is more general and can be used t o examine how different implementations respond dynamically to dependent sequences of insertion and deletion requests. The Markov hold model is used t o conduct tests based on random, stressed, and correlated input sequences of requests, with performance measures including completion times, sensitivity t o correlations, sensitivity to duplication, and eficiency of data-handling. W e apply these tests t o fourteen different simulation calendar implementations.