Computer simulations have been used more than ever before to embark on developing and understanding complex systems such as Multi-Agent Systems (MAS). As simulation systems become larger and more complex, scalability becomes an important issue for their successful deployment. PDES-MAS (Parallel and Discrete Event Simulations for Multi-Agent Systems) framework is, implemented based on distributed shared memory architecture, a parallel and discrete event simulation kernel to distribute and run parallel simulation of Multi-Agent Systems (MAS). There are several issues within such system and this thesis presents a notion of logical time synchronised range queries to address the issue of data access. Accessing data efficiently in a latency-sensitive and large scale network overlay is a vital requirement for the scalability of the system. So, this thesis presents a notion of synchronised range queries with algorithms to manage distributed data structures consistently and in a time ordered fashion across the system. To localise data access in such a large scale simulation system, algorithms are provided to distribute shared state such that the distribution reflects access patterns of simulating nodes. The algorithms are evaluated within the implementation of PDES-MAS framework using various agent based simulation traces.
展开▼
