Abstract
Cluster systems are widely used in modern high performance computing. With the rapidly increasing of parallel algorithms, it is an open problem to analyze and evaluate whether they take good advantage of the computing and network resources of clusters.[1 − 3] We present a novel mathematic model(n-Cube Model for Cluster Computing) that epitomizes the algorithms commonly used on clusters and evaluate this model using Stochastic Petri Nets (SPN). The state space of our model’s SPN is also discussed formally. Finally, we take MM5(the Fifth- Generation Model) as a case and the comparative performance analysis shows the immense vitality of the model.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Koibuchi, M., Watanabe, K., Kono, K., Jouraku, A., Amano, H.: Performance evaluation of routing algorithms in RHiNET-2 cluster. In: Cluster Computing. Proceedings 2003 IEEE International Conference, pp. 395–402 (2003)
Bessonov, O., Fougere, D., Roux, B.: Using a parallel cfd code for evaluation of clusters and MPPs. In: Parallel and Distributed Processing Symposium, Proceedings 2003 International, pp. 65–72 (2003)
Nguyen, K.N., Le, T.T.: Evaluation and comparison performance of various MPI implementations on an OSCAR linux cluster. In: Information Technology: Coding and Computing [Computers and Communications]. Proceedings. ITCC 2003 International Conference, pp. 310–314 (2003)
Xu, K., Fan, X.-b., Lin, C., Wu, J.-p.: Performance model and analysis of a distributed router. In: Communications, Circuits and Systems. IEEE 2002 International Conference, vol. 1, pp. 786–790 (2002)
Lin, C.: Performance evaluation of the computer network and computer system. Tsinghua University Press (2001)
Lin, C., Marinescu, D.C.: Stochastic high-level Petri nets and applications. IEEE Transactions on Computers 37(7), 815–825 (1988)
Lin, C., Qu, Y., Ren, F., Marinescu, D.C.: Performance Equivalent Analysis of Workflow Systems Based on Stochastic Petri Net Models. In: Proceedings of the First International Conference on Engineering and Deployment of Cooperative Information Systems (2002)
Molloy, M.K.: Performance Analysis Using Stochastic Petri Nets. IEEE Trans. Comp. C-39(9), 913–917 (1982)
Lopez-Benitez, N.: Dependability analysis of distributed computing systems using stochastic Petri nets. Reliable Distributed Systems, pp. 85 - 92 (1992)
Grell, G.A., Dudhia, J., Stauffer, D.R.: A Description of the Fifth-Generation Penn State/NCAR Mesoscale Model (MM5). NCAR Technical Note NCAR/TN-398+STR, (June 1994)
Michalakes, J: The same source parallel MM5. In: Proceedings Second International Workshop on Software Engineering and Code Design in Parallel Meteorological and Oceanographic Applications 1998, Greenbelt, MD, USA.
MM5 Homepage, http://www.mmm.ucar.edu/mm5/mm5-home.html
Ming-Yun, H.: Analysis of Boolean N-cube interconnection networks for multiprocessor systems. Doctoral Dissertation
Almeida, V.A.F., Dowdy, L.W., Leuze, M.R.: An analytic model for parallel Gaussian elimination on a binary N-Cube architecture. In: Proceedings of the third conference on Hypercube concurrent computers and applications, pp. 1550 - 1553 (1989)
Yang, C.S., Wang, J.F., Lee, J.Y., Boesch, F.T.: Graphic Theoretic Reliability Analysis for the Boolean n cube networks. IEEE Transactions on circuits and systems 35(9), 1175–1179 (1988)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Song, T., Wang, D., Hu, M., Xue, Y. (2007). n-Cube Model for Cluster Computing and Its Evaluation. In: Xu, M., Zhan, Y., Cao, J., Liu, Y. (eds) Advanced Parallel Processing Technologies. APPT 2007. Lecture Notes in Computer Science, vol 4847. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-76837-1_38
Download citation
DOI: https://doi.org/10.1007/978-3-540-76837-1_38
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-76836-4
Online ISBN: 978-3-540-76837-1
eBook Packages: Computer ScienceComputer Science (R0)