MCMAS: A toolkit for developing agent-based simulations on many-core architectures

Guillaume Laville (1) , Christophe Lang (1) , Bénédicte Herrmann (1) , Laurent Philippe (1) , Kamel Mazouzi (2) , Nicolas Marilleau (3)

1 FEMTO-ST - Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174)
2 UFC - Université de Franche-Comté
3 UPMC - Université Pierre et Marie Curie - Paris 6

Multi-agent models and simulations are used to describe complex systems in domains such as biological, geographical or ecological sciences. The increasing model complexity results in a growing need for computing resources and motivates the use of new architectures such as multi-cores and many-cores. Using them efficiently however remains a challenge in many models as it requires adaptations tailored to each program, using low-level code and libraries. In this paper we present MCMAS a generic toolkit allowing an efficient use of many-core architectures through already defined data structures and kernels. The toolkit provides few famous algorithms as diffusion, path-finding or population dynamics that are frequently used in an agent based models. For further needs, MCMAS is based on a flexible architecture that can easily be enriched by new algorithms thanks to development features. The use of the library is illustrated with three models and their performance analysis.

Domaines: Cryptographie et sécurité [cs.CR]; Calcul parallèle, distribué et partagé [cs.DC]; Technologies Émergeantes [cs.ET]; Informatique ubiquitaire; Système multi-agents [cs.MA]; Modélisation et simulation; Génie logiciel [cs.SE]

Guillaume Laville, Christophe Lang, Bénédicte Herrmann, Laurent Philippe, Kamel Mazouzi, et al.. MCMAS: A toolkit for developing agent-based simulations on many-core architectures. Multiagent and Grid Systems, 2015, 11 (1), pp.15 - 31. ⟨hal-02131113⟩