Efficient Agent-Based Simulations Using the Sender Asynchronous Programming Model

Abstract

This paper develops and evaluates an Agent-Based Model (ABM) simulator for disease transmission using the C++ Sender Programming Model. The asynchronous approach with the Sender Model is, on average, 2.61 times faster than synchronous methods, enhancing performance while preserving cross-platform compatibility. This research offers a valuable alternative for simulating complex epidemiological scenarios, advancing computational epidemiology by optimizing both performance and portability. Future work will focus on improving memory management and validating the model across different hardware configurations and population densities.