Performance Prediction of Multi-Agent Based Simulation Applications on the Grid

Authors: Dawit Mengistu, Lars Lundberg, Paul Davidsson

Abstract:

A major requirement for Grid application developers is ensuring performance and scalability of their applications. Predicting the performance of an application demands understanding its specific features. This paper discusses performance modeling and prediction of multi-agent based simulation (MABS) applications on the Grid. An experiment conducted using a synthetic MABS workload explains the key features to be included in the performance model. The results obtained from the experiment show that the prediction model developed for the synthetic workload can be used as a guideline to understand to estimate the performance characteristics of real world simulation applications.

Keywords: Grid computing, Performance modeling, Performance prediction, Multi-agent simulation.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1059974

References:

[1] Cioffi-Revilla, C. "Invariance and universality in social.agent-base simulations," Proc. National Academy of Science USA, 99 (2002) Suppl. 3: 7314-6
[2] Davidsson, P. et al. "Applications of multi-agent based simulations," Seventh Workshop on Multi-Agent-Based Simulation (2006), Future University-Hakodate, Japan.
[3] Sansores, C. and Pavon, J. "A framework for agent based social simulation," The Second European Workshop on Multi-Agent Systems (2004), Barcelona, Spain.
[4] Gasser, L. "Smooth scaling ahead: Progressive MAS simulation from single PCs to Grids," Joint Workshop on Multi-Agent and Multi-Agent- Based Simulation (2004) New York
[5] Ferreira L. et. al. "The IBM Red Book. Introduction to Grid Computing with Globus", (2003).
[6] Sotomajor, B. "The Globus toolkit 4 Programmer-s Tutorial", (2005).
[7] Helsinger, A. et al. "Tools and techniques for performance measurement of large distributed multi-agent systems," AAMAS-03 (2003) Australia
[8] Xu, Z., Miller, B.P. and Naim, O. "Dynamic instrumentation of threaded applications," Proc. 7th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (1999), Georgia USA.
[9] Tirado-Ramos, A., Groen D., Sloot, P. "On-line Application Performance Monitoring of Blood Flow Simulation in Computational Grid Architectures," Proceedings of the 18th IEEE Symposium on Computer-Based Medical Systems, (2005)
[10] Iskra, K.A., Albada, G.D., Sloot, P.M.A. "Towards Grid-Aware Time Warp", Proceedingsof the 18th Workshop on Parallel and Distributed Simulation, (2004)
[11] Puppin D., Tonellotto N., and Laforenza D. "Using Web Services to Run Distributed Numerical Applications", in LNCS vol. 3241 D.Krazmuller Springer-verlag Berlin Heidelberg (2004) pp. 207-214.
[12] Timm I.J., Pawlaszczyk D., "Large Scale Multiagent Simulation on the Grid," Proceedings of 5th IEEE International Symposium on Cluster Computing and the Grid. IEEE Computer Society Washington, DC, USA, (2005).
[13] Barnett J., "The behaviour of Java threads under Linux NPTL" , (2003).
[14] Tichy, W.F. "Should computer scientists experiment more?," IEEE Computer, USA. Vol. 31 No. 5 (1998), pp.32-40
[15] Jarvis, S.A. Spooner, D.P. Keung, H.N.L.C. Nudd, G.R. "Performance prediction and its use in parallel and distributed computing systems," Proceedings of the 17th International Symposium on Parallel and Distributed Processing (2003) .IEEE Computer Society Washington, DC, USA
[16] Badia, Rosa M., "Performance Prediction in a Grid Environment," 1st European Across Grids Conference, Santiago de Compostela July 2003.
[17] Jarvis, S.A. et. al. "performance-responsive middleware for grid computing", Proceedings of UK e-Science All Hands Meeting, (2003) Nottingham, UK.
[18] Collis J., Ndmu D., Buskirk C., "The Zeus Agent Building Toolkit", (2000).