A Scalable Architecture for Crowd Simulation: Implementing a Parallel Action Server

  title={A Scalable Architecture for Crowd Simulation: Implementing a Parallel Action Server},
  author={Guillermo Vigueras and Miguel Lozano and Carlos Perez and Juan M. Ordu{\~n}a},
  journal={2008 37th International Conference on Parallel Processing},
Crowd simulation can be considered as a special case of virtual environments where avatars are intelligent agents instead of user-driven entities. These applications require both rendering visually plausible images of the virtual world and managing the behavior of autonomous agents. Although several proposals have focused on the software architectures for these systems, the scalability of crowd simulation is still an open issue. In this paper, we propose a scalable architecture that can manage… 

Figures from this paper

A scalable multiagent system architecture for interactive applications

A distributed visualization system for crowd simulations

This paper proposes a distributed visualization system for large crowds of autonomous agents that allows the visualization of crowds animated with complex behaviors without adding significant overhead to the simulation servers, and it takes advantage of the Graphics Processor Unit GPU capabilities.

A GPU-Based Multi-agent System for Real-Time Simulations

This paper proposes a GPU-based multi-agent system for crowd simulation that can support agents with complex navigational behaviors and significantly increases the parallelism achieved with respect to the implementation of the same distributed server on a CPU.

Performance improvements of real-time crowd simulations

  • G. ViguerasJ. OrduñaM. Lozano
  • Computer Science
    2010 IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum (IPDPSW)
  • 2010
A distributed software architecture that can take advantage of the existing distributed and multi-core architectures and allow the use of GPUs not only for rendering images but also for computing purposes is proposed.

Improving the Performance of Partitioning Methods for Crowd Simulations

This paper proposes the improvement of the partitioning method for distributed crowd simulations by using irregular shape regions, and proposes the partition of the virtual world using convex hulls.

A Partitioning Method based on Convex Hulls for Crowd Simulations

The performance evaluation results show that the Convex Hull method outperforms the rest of the considered methods in terms of both fitness function values and execution times, regardless of the movement pattern followed by the agents.

A comparative study of partitioning methods for crowd simulations

Workload balancing in distributed crowd simulations: the partitioning method

This paper proposes the integration of a workload balancing technique for crowd simulations within a partitioning method based on convex hulls, and shows that this technique ensures the saturation avoidance of the servers in an homogeneous distributed system.

Accelerating collision detection for large-scale crowd simulation on multi-core and many-core architectures

This paper proposes different parallelization strategies for the collision check procedure that takes place in agent-based simulations, and proposes a parallel CPU version that uses read--copy update (RCU), a new synchronization method which significantly improves performance.

A data parallel approach to modelling and simulation of large crowd

In this study, a method was proposed to formulate crowd behavior with the cell automata and multi-agent models, which were successfully mapped onto the MapReduce programming model to simulate large crowd scenarios over a cluster.



On the Design of an Efficient Architecture for Supporting Large Crowds of Autonomous Agents

This proposal consists of a cluster of computers in order to improve flexibility and robustness, as well as a hierarchical software architecture that efficiently provides consistency that allows to efficiently manage thousands of autonomous agents.

Scalable behaviors for crowd simulation

This paper presents a new approach to controlling the behavior of agents in a crowd that is scalable in the sense that increasingly complex crowd behaviors can be created without a corresponding increase in the complexity of the agents.

Controlling individual agents in high-density crowd simulation

The HiDAC system (for High-Density Autonomous Crowds) focuses on the problem of simulating the local motion and global wayfinding behaviors of crowds moving in a natural manner within dynamically changing virtual environments.

Big fast crowds on PS3

An implementation of a scalable multi-processor approach to large, fast crowd simulations, as in Quinn et al. 2003, is described for PLAYSTATION®3 which supports simulation and display of simple crowds of up to 15,000 individuals at 60 frames per second.

Networked Virtual Environments

The familiarization the end users with the Internet and the promising advantages and opportunities of Virtual Reality contributed to currently view NVEs as an effective tool for supporting communication and collaboration of scattered users.


It is demonstrated how the use of a multicomputer can greatly accelerate the speed of a pedestrian movement simulator based on the social forces model and the parallel algorithm is highly scalable, meaning that adding processors will enable the simulation of even larger crowds.

Partitioning crowded virtual environments

This work investigates several techniques that partition a crowded virtual environment into regions that can be managed by separate servers or mapped onto different multicast groups, and finds that the region growing technique give the best overall results.

Flocks, herds and schools: A distributed behavioral model

This paper explores an approach based on simulation as an alternative to scripting the paths of each bird individually, an elaboration of a particle systems, with the simulated birds being the particles.

Autonomous pedestrians

This comprehensive model integrates motor, perceptual, behavioral, and cognitive components within a model of pedestrians as individuals, yielding results of unprecedented fidelity and complexity for fully autonomous multi-human simulation in a large urban environment.

Parallel simulation of group behaviors

  • Bo ZhouSuiping Zhou
  • Computer Science
    Proceedings of the 2004 Winter Simulation Conference, 2004.
  • 2004
The proposed parallel algorithm provides good speedup in generating flocking behaviors compared with the sequential simulation and the new partitioning and communication mechanisms in the parallel algorithm make the flocking simulation more efficient.