Alessandro Fabbri
Dipartimento di Scienze dell'Informazione
Università di Bologna
Mura Anteo Zamboni 7
I - 40127 Bologna (Italy)
email: fabbria@cs.unibo.it
We present State Query Time Warp (SQTW), a parallel discrete-event simulation policy for simulation of state-dependent performance models. SQTW integrates Time Warp, an optimistic policy, with a mechanism whereby state-dependent evolutions are allowed. A SQTW-based simulator was implemented, and experimental measures concerning SQTW overhead and efficiency assessments are reported. Results point out SQTW overheads with respect to Time Warp, but also show high state computation efficiencies and possible speed-up achievements.