ABSTRACT
Construction of prognoses about environmental changes demands simulations of massive multi-agent models. This work evaluates the hypothesis that the combined use of techniques such as annotation and bag of tasks can result in flexible and scalable platforms for multi-agent simulation. Although these are well known techniques, most environmental modeling platforms use other approaches to provide high performance computing. In general, the approach used is dependent of the modeling paradigm theses platforms implement. We are looking for approaches that can cope with multiple modeling paradigms. To evaluate our hypothesis, the TerraME modeling platform was extended to run over SMPs (Symmetric Multiprocessors) architectures and used in real case studies. While annotation allows modelers to implement different parallelization strategies without prevent models to run over sequential architectures, the bag of tasks provides load balancing over multiprocessors. The results demonstrated that 35% of linear speedup can be obtained for models with high dependence among tasks, when 8 processors are used. Moreover, for models that have low data or control dependencies, around 90% of linear speedup can be obtained.
- Câmara G., Vinhas L., et.al. TerraLib: An Open Source GIS Library for Large-Scale Environmental and Socio-Economic Applications, Open Source Approaches in Spatial Data Handling, pages 247--270. 2008.Google ScholarCross Ref
- Carneiro, T. G. S. Nested-CA: A foundation for multiscale modeling of land use and land change. phd thesis in computer science, The National Institute for Space Research, Brazil, 2006.Google Scholar
- Chave, J. Study of structural, sucessional and spatial patterns in tropical rain forest using troll, a spatially explicit forest model. Ecological Modeling, 1999.Google Scholar
- Collier N. Repast HPC Manual. Argonne National Laboratory, Argone, IL, USA, 2012.Google Scholar
- Collier N. and North M., Repast HPC: A platform for Large-Scale Agent-Based Modeling. Computing Techniques for Complex System Simulation, November 29, 201 ISBN-13: 978-0470592441.Google Scholar
- Cordasco G., Chiara D. R., et al. A Framework for distributing Agent-based simulations. EURO PAR 2011: PARALLEL PROCESSING WORKSHOPS, Lecture Notes in Computer Science, 2012, Volume: 7155/2012, 460--470, DOI: 10.1007/978-3-642-29737-3_51. Google ScholarDigital Library
- Filippi, J., and Bisgambiglia, P.. JDEVS: an implementation of a DEVS based formal framework for environmental modelling. Environmental Modelling and Software 19(3) 261--274. 2004.Google ScholarCross Ref
- Ierusalimschy, R, Programming in Lua, Lua.Org, ISBN 85-903798-1-7. 2003.Google Scholar
- Luke S. Multiagent Simulation And the MASON Library. George Mason University, First Edition, August, 2011.Google Scholar
- Luke S., Cioffi-Revilla C., et al., MASON: A New Multi-Agent Simulation Toolkit. Proceedings of the 2004 Swarmfest Workshop. 2004.Google Scholar
- Lottiaux, R. ; Vallee, G. ; et al., Kerrighed and data parallelism: cluster computing on single system image operating systems. Proceedings of the International Conference on Cluster Computing, 2004. Google ScholarDigital Library
- Minar, N., Burkhart, R., Langton, C., Askenazi , M.,. The Swarm Simulation System: A Toolkit for Building Multi-Agent Simulation. SFI Working Paper 96-06-042. 1996.Google Scholar
- Shahid Hussain and Hassan Shabbir. Directory scalability in multi-agent based systems. Master's thesis in Science in Software Engineering, School of Engineering at Blekinge Institute of Technology.2008Google Scholar
- Tisue, S., and Wilensky, U.. NetLogo: A Simple Environment for Modeling Complexity, International Conference on Complex Systems: Boston. 2004.Google Scholar
Index Terms
- TerraME HPA: parallel simulation of multi-agent systems over SMPs
Recommendations
Wisconsin Wind Tunnel II: A Fast, Portable Parallel Architecture Simulator
Analysis of future parallel computers requires rapidly simulating target designs running realistic workloads. Two techniques have accelerated such simulations: direct execution and using a parallel host. Historically, these techniques have lacked ...
Accurate Functional-First Multicore Simulators
Fast and accurate simulation of multicore systems requires a parallelized simulator. This paper describes a novel method to build cycle-accurate-capable and parallelizable functional-first simulators of multicore targets.
Comments