In Tolk, A., Diallo, S. D., Ryzhov, I. O., Yilmaz, L., Buckley, S., & Miller, J. A., editors, Proceedings of the 2014 Winter Simulation Conference, pages 3761–3772, Piscataway, NJ, 2014. Institute of Electrical and Electronics Engineers, Inc.. Paper doi abstract bibtex
Ranking and selection (R&S) procedures designed for serial computing environments include two-stage procedures (e.g., NSGS), which provide guarantees on good selection and are simple to implement, and fully-sequential screening-based procedures (e.g., KN), which save simulation replications by eliminating inferior systems early. In a parallel computing environment, a naively-parallelized NSGS procedure may require more simulation replications than a sequential screening procedure such as that of Ni, Hunter, and Henderson (2013) (NHH), but requires less communication since there is no periodic screening. The parallel procedure NHH may require less simulation replications overall, but requires more communication to implement periodic screening. We numerically explore the tradeoffs between these two procedures on a parallel computing platform. In particular, we discuss their statistical validity, efficiency, and implementation, including communication and load-balancing.