What is it about?
This article compared three C/C++ cluster-computing libraries for parallelizing agent-based models: FLAME, RepastHPC, and MASS C++. The comparison was conducted from programmability and parallel performance viewpoints, using seven benchmark programs.
Featured Image
Photo by Fotis Fotopoulos on Unsplash
Why is it important?
It is the very first article to empirically compare C/C++ cluster-computing libraries for parallelizing agent-based models, using seven benchmark programs that cover social, behavioral, economic, biological and city-planning application domains.
Perspectives
MASS C++, one of the three libraries we compared, is the one we have developed from scratch. Our intent was to justify its strength and challenges in comparison with two major parallel libraries: FLAME from UK and RepastHPC from US. All the work has been done with 1 master's student and 4 undergraduate students under the supervision of their faculty member, (i.e., the 1st author).
Munehiro Fukuda
University of Washington Bothell
Read the Original
This page is a summary of: An Analysis of Three C/C++ Cluster-Computing Libraries for Agent-Based Models, ACM Transactions on Modeling and Computer Simulation, May 2025, ACM (Association for Computing Machinery),
DOI: 10.1145/3732778.
You can read the full text:
Contributors
The following have contributed to this page
