Antoine Dutot

Organization detection for dynamic load balancing in individual-based simulations

Large-scale individual-based simulations can benefit a lot from high performance computing environments. The benefit that can be hopped for depends greatly on a good load distribution among the processing resources together with the minimization of the communication overhead. However, minimizing both idle time and communication overhead requires the search for a trade-off. Inspired by complex systems, the approach described in this paper aims at minimizing the volume of data exchanged over the network between tasks of a distributed application, while balancing the load between available computing ressources. The method lies on the trail-laying trail-following paradigm used in algorithms based on artificial ants.

Dynamic placement using ants for object based simulations

A distributed application may be considered as a set of interacting entities continuously evolving. Such application can be modeled as a graph with one-to-one mappings between vertices and entities and between edges and communications. Performances depend directly on a good load balancing of the entities between available computing devices and on the minimization of the impact of the communications between them. However, both objectives are contradictory and good performances are achieved if and only if a good tradeoff is found. Our method for finding such a tradeoff is new and based on colored ant colonies. Each computing resource is associated to one ant colony characterized by a color, allowing an implicit consideration of the load balancing constraint. Then, using colored pheromones, ants are just seeking for communicating structures. The method operates on graphs which structural and numerical parameters may change dynamically during the execution.

Communities detection algorithm to minimize risk during an evacuation

Recent dramatic events recall to the world that is has to deal with risk problematic. Thus, to face risk in an agglomeration, we study hazards from natural or anthropic origin. One problem is to decide if it is necessary to evacuate or confine population. To help decision makers, we analyze the road network structure which may influence flow fluidity especially in a dangerous case. In this work, we detail an algorithm to detect communities in large graphs. It allows to identify routes that may cause problems in an evacuation case. Thanks to this algorithm, we study a toxic cloud propagation in a given zone and identify roads to avoid when evacuating this zone

Competing Ants for Organization Detection Application to Dynamic Distribution

A simulation application may be modeled as a set of interacting entities within an environment. Such applications can be represented as a graph with a one-to-one mapping between vertices and entities and between edges and communications. As for classical applications, performances depend directly on a good load balancing of the entities between available computing devices and on the minimization of the impact of the communications between them. However, both objectives are contradictory and good performances may be achieved if and only if a good trade off is found. Our method for finding such a trade off leans on a bio-inspired method. We use competitive colonies of numerical ants, each one depositing colored pheromones, to find organizations of highly communicating entities.

Colored Ants for Distributed Simulations

Complex system simulations can often be represented by an evolving graph which evolves with a one-to-one mapping between vertices and entities and between edges and communications. Performances depend directly on a good load balancing of the entities between available computing devices and on the minimization of the impact of the communications between them. We use competing colonies of numerical ants, each depositing distinctly colored pheromones, to find clusters of highly communicating entities. Ants are attracted by communications and their own colored pheromones, while repulsion interactions between colonies allow to preserve a good distribution.

TL a Language to Create Games for Visually Impaired Children

In this paper, we present TL, a language for creating games for visually impaired and blind children [1]. This language is a part of the TiM project whose overall aim is to offer to young visually impaired children the possibility to play with computer games.

Agent-based Approach and Dynamic Graphs to Model Logistic Corridor

This paper presents the modelling of a logistic corridor. It integrates the port and the metropolitan logistics connected by an interface. Such a system can be seen as complex. A multi-scale point of view is adopted thanks to an agent-based approach which is coupled with dynamic graphs in order to represent in the one hand the actors involved in the transportation of goods, and in the other hand, the structured environment. The model is implemented in an agent-based simulation platform. Results about the impacts of parameters on the demand generating the flows of goods are finally discussed.

A Model of Road Network and Buildings Extension Co-evolution☆

Abstract Urban morphology tries to understand the spatial structure of cities. It searches to identify the patterns and underlying substructures but also the process of city development. Our contribution follows this last direction, we model the co-evolution of the road network and the buildings. Our goal is triple: to combine cellular automata simplicity with city irregularities, to enable the road network and buildings co-evolution and to respect temporal coherence. These objectives lead us to propose a simulation model based on a geographic automata system. Several kinds of models already exist, but they usually concentrate only on the evolution of the road network or only on the population density. The originality of our work consists in the dynamic of co-evolution of the city, we propose a city morphogenesis model.

4 – Networking, Networks and Dynamic Graphs

Networks cannot be confined within a single field of research. The concepts associated with them have considerably changed over time, and continue to do so today. This makes it difficult to give a single clear definition of a network. Networks are also the location of processes and so are embedded in time, and their structure can change; these are the observations on which our approach is based. Throughout this chapter, we will attempt to shed light on these scientific objects by accepting the bias inherently present in the desire to produce mathematical and computational models for studying networks with NetLogo.

5 – Swarm Problem-Solving

: It is increasingly common for algorithms in computer science to be inspired by “natural” models. This is not a new trend. Computer science has always drawn from its surroundings as a source of inspiration and our user interfaces are proof of this. Examples of algorithms and programming models like this include, among others, simulated annealing, cellular automata, DNA computing, evolutionary algorithms and artificial chemistry.