Dynamic cancellation: Selecting time warp cancellation strategies at runtime

The performance of Time Warp parallel discrete event simulators can be affe cted by the cancellation strategy used to send anti-messages. Under aggress ive cancellation, anti-message generation occurs immediately after a stragg ler message is detected. In contrast, lazy cancellation delays the sending of anti-messages until forward processing from a straggler message confirms that the premature computation did indeed generate an incorrect message. P revious studies have shown that neither approach is clearly superior to the other in all cases (even within the same application domain). Furthermore, no strategy exists to make a priori determination of the more favorable ca ncellation strategy. Most existing Time Warp systems merely provide a switc h for the user to select the cancellation strategy employed. This paper exp lores the use of simulation time decision procedures to select cancellation strategies. The approach is termed Dynamic Cancellation and it assigns the capability for selecting cancellation strategies to the Logical Processes (LPs) in a Time Warp simulation. Thus, within a single parallel simulation both strategies may be employed by distinct LPs and even across the simulat ion lifetime of an LP. Empirical analysis using several control strategies show that dynamic cancellation always performs with the best static strateg y and, in some cases, dynamic cancellation provides some nominal (5-10%) pe rformance gain over the best static strategy.