Ton Oguara

An adaptive load management mechanism for distributed simulation of multi-agent systems

The paper presents a load management mechanism for distributed simulations of multi-agent systems. The mechanism minimizes the cost of accessing the shared state in the distributed simulation by dynamically redistributing shared state variables according to the access pattern of the simulation model. To evaluate the effectiveness and performance of the mechanism, a series of benchmark experiments were performed using the PDES-MAS framework for distributed simulation of multi-agent systems. Although preliminary, the results indicate that the proposed mechanism significantly reduces the overall access cost of the system.

Decision-theoretic throttling for optimistic simulations of multi-agent systems

In this paper we present a throttling mechanism for optimistic simulations of multi-agent systems, which delays read accesses to the shared simulation state that are likely to be rolled back. We develop a decision-theoretic model of rollback and show how this can be used to derive the optimal time to delay a read event so as to minimize the expected overall execution time of the simulation. We briefly describe an implementation of this approach in ASSK, a distributed simulation kernel developed to investigate synchronization mechanisms for MAS simulation, and report the results of preliminary experiments to evaluate the effectiveness of our approach.

Data access in distributed simulations of multi-agent systems

Distributed simulation has emerged as an important instrument for studying large-scale complex systems. Such systems inherently consist of a large number of components, which operate in a large shared state space interacting with it in highly dynamic and unpredictable ways. Optimising access to the shared state space is crucial for achieving efficient simulation executions. Data accesses may take two forms: locating data according to a set of attribute value ranges (range query) or locating a particular state variable from the given identifier (ID query and update). This paper proposes two alternative routing approaches, namely the address-based approach, which locates data according to their address information, and the range-based approach, whose operation is based on looking up attribute value range information along the paths to the destinations. The two algorithms are discussed and analysed in the context of PDES-MAS, a framework for the distributed simulation of multi-agent systems, which uses a hierarchical infrastructure to manage the shared state space. The paper introduces a generic meta-simulation framework which is used to perform a quantitative comparative analysis of the proposed algorithms under various circumstances.

HLA_AGENT: Distributed Simulation of Agent-Based Systems with HLA

In this paper we describe hla_agent, a tool for the distributed simulation of agent-based systems, which integrates the sim_agent agent toolkit and the High Level Architecture (HLA) simulator interoperability framework. Using a simple Tileworld scenario as an example, we show how the HLA can be used to flexibly distribute a sim_agent simulation with different agents being simulated on different machines. The distribution is transparent in the sense that the existing sim_agent code runs unmodified and the agents are unaware that other parts of the simulation are running remotely. We present some preliminary experimental results which illustrate the performance of hla_agent on a Linux cluster running a distributed version of Tileworld and compare this with the original (non-distributed) sim_agent version.

Analysing Probabilistically Constrained Optimism

In previous work we presented the DTRD algorithm, an optimistic synchronisation algorithm for parallel discrete event simulation of multi-agent systems, and showed that it outperforms Time Warp and time windows on range of test cases. DTRD uses a decision theoretic model of rollback to derive an optimal time to delay read event so as to maximise the rate of LVT progression. The algorithm assumes that the inter-arrival times (both virtual and real) of events are normally distributed. In this paper we present a more detailed evaluation of the DTRD algorithm, and specifically how the performance of the algorithm is affected when the inter-arrival times do not follow the assumed distributions. Our analysis suggests that the performance of the algorithm is relatively insensitive to events whose inter-arrival times are not normally distributed. However as the variance of the input events increases its performance degrades to that of Time Warp. Our approach to evaluation is general, and we outline how the analysis may be applied to other decision theoretic algorithms.

Analysing the Performance of Optimistic Synchronisation Algorithms in Simulations of Multi-Agent Systems

In this paper we present a detailed analysis of the performance of the Decision Theoretic Read Delay (DTRD) optimistic synchronisation algorithm for simulations of Multi- Agent Systems. We develop an abstract characterisation of the access patterns found in MAS simulations based on the simulation’s degree of coupling and skew. Using this characterisation, we generated stereotypical test cases which we used to compare the performance of the DTRD algorithm with that of Time Warp and time windows. To determine if the test cases reliably predict performance in a real agent simulation, we compared the predictions made by the test cases with performance results from the Boids agent simulation benchmark for a range of simulation parameters. The results indicate that DTRD adapts to the mixtures of coupling cases found in real agent simulations and is capable of tracking changes in coupling during the simulation.

Performance Analysis of Shared Data Access Algorithms for Distributed Simulation of Multi-Agent Systems

Distributed simulation is an important instrument for studying multi-agent systems (MAS). Such large scale MAS simulations often have a large shared state space. Moreover, the shared state and the access pattern of agent simulations both are highly dynamic and unpredictable. Optimising access to the shared data is crucial for achieving efficient simulation executions. PDES-MAS is a framework for distributed simulation of MAS, which uses a hierarchical infrastructure to manage the shared data. In order to enable agent simulations to access distributed shared data efficiently, this paper proposes two routing algorithms, namely the address-based routing and the range-based routing. The paper introduces a meta-simulation approach to evaluate the characteristics of both solutions and provides a quantitative comparative analysis of the proposed algorithms.