Publications with the Performability Engineering Research Group
Adaptive Uniformization: Technical Details.
A. P. A. van Moorsel and W. H. Sanders. (93M03.pdf)
Research Report 93M03 CRHC.
Technical details complementing the following paper:
Adaptive Uniformization.
A. P. A. van Moorsel and W. H. Sanders. (94M02.pdf)
ORSA Communications in Statistics: Stochastic Models, vol. 10, no. 3, August 1994, pp. 619-648.
Discusses an efficient variation of uniformization, particularly useful for stiff systems. A formal definition of adaptive uniformization is given, along with a proof that it yields correct results, and examples of its use.
Algorithms for the Generation of State-Level Representations of Stochastic Activity Networks with General Reward Structures.
M. A. Qureshi, W. H. Sanders, A. P. A. van Moorsel, and R. German. (95Q02.pdf)
Proceedings of the Sixth International Workshop on Petri Nets and Performance Models, Durham, NC, October 3-6, 1995, pp. 180-190.
This paper discusses the generation of the stochastic process underlying a SAN, including the algorithm for the well-specified check. Furthermore, a general reward structure is introduced that can represent all reward variables defined on the marking behavior of a SAN.
UltraSAN Version 3 Overview.
D. D. Deavours, W. D. Obal II, M. A. Qureshi, W. H. Sanders, and A. P. A. van Moorsel. (95DEA01.pdf)
Proceedings of the Sixth International Workshop on Petri Nets and Performance Models, Durham, NC, October 3-6, 1995, pp. 216-217.
A two-page overview of UltraSAN Version 3.0.
Computation of the Asymptotic Bias and Variance for Simulation of Markov Reward Models.
A. P. A. van Moorsel, L. A. Kant, and W. H. Sanders. (96MOO01.pdf)
Proceedings of the 29th Annual Simulation Symposium, New Orleans, LA, April 1996, pp. 173-182.
Gives a numerical method for computing the asymptotic variance for large Markov models. The asymptotic bias and variance determine the goodness and run length of a simulation.
Expected Impulse Rewards in Markov Regenerative Stochastic Petri Nets.
R. German, A. P. A. van Moorsel, M. A. Qureshi, and W. H. Sanders. (95GER01.pdf)
Application and Theory of Petri Nets, Proceedings of the 17th International Conference, Osaka, Japan, June 24-28, 1996, pp. 172-191 (ed. J. Billington and W. Reisig), Lecture Notes in Computer Science, Vol. 1091, Springer-Verlag, 1996.
Computes reward measures for Petri nets with general distributions. Furthermore, the method of supplementary variables is enhanced to derive transient solutions of models with general distributions.
Algorithms for the Generation of State-Level Representations of Stochastic Activity Networks with General Reward Structures.
M. A. Qureshi, W. H. Sanders, A. P. A. van Moorsel, and R. German. (95QUR05.pdf)
IEEE Transactions on Software Engineering, vol. 22, no. 9, September 1996, pp. 603-614.
Same description as for conference version, above.
Transient Solution of Markov Models by Combining Adaptive & Standard Uniformization.
A. P. A. van Moorsel and W. H. Sanders. (96MOO02.pdf; file is of a prepublication version of this paper)
IEEE Transactions on Reliability, vol. 46, no. 3, September 1997, pp. 430-440.
Proposes a new transient solution method that combines adaptive and standard uniformization in a method that exhibits the benefits of both methods and, to a large extent, eliminates their disadvantages.
|
