The inherent variability of multithreaded commercial workloads can lead to incorrect results in architectural simulation studies. Random perturbations and statistical methods help computer architects draw valid conclusions.
1. L.A. Barroso, K. Gharachorloo,, and E. Bugnion., "Memory System Characterization of Commercial Workloads," Proc. 25th Int'l Symp. on Computer Architecture (ISCA 98), pp. 3-14.
2. A.R. Alameldeen, and D.A. Wood,, "Variability in Architectural Simulations of Multi-threaded Workloads," Proc. 9th Int'l Symp. on High-Performance Computer Architecture (HPCA 03), pp. 7-18.
3. T. Sherwood,, et al., "Automatically Characterizing Large Scale Program Behavior," Proc. 10th Int'l Conf. on Architectural Support for Programming Languages and Operating
Systems (ASPLOS-X), 2002, pp. 45-57.
4. A.R. Alameldeen, et al., "Simulating a $2M Commercial Server on a $2K PC," Computer, vol. 36, no. 2, Feb. 2003, pp. 50-57.
5. C.J. Mauer, M.D. Hill,, and D.A. Wood,, "Full System Timing-First Simulation," Proc. 2002 ACM Sigmetrics Conf. on Measurement and Modeling of Computer Systems, ACM Press, 2002, pp. 108-116.
6. H. Frank, and S.G. Althoen,, Statistics: Concepts and Applications, Cambridge Univ. Press, first edition, 1994, pp. 274-452.