The analysis of real systems and the development of predictive models to describe the evolution of real phenomena are challenging tasks that can improve the design of methodologies in many research fields. In this context, Agent-Based Model (ABM) can be seen as an innovative tool for modelling real-world complex simulations. This paper presents Rust-AB, an open-source library for developing ABM simulation on sequential and/or parallel computing platforms, exploiting Rust as programming language. The Rust-AB architecture as well as an investigation on the ability of Rust to develop ABM simulations are discussed. An ABM simulation written in Rust-AB, and a performance comparison against the well-adopted Java ABM toolkit MASON is also presented.
On Evaluating Rust as a Programming Language for the Future of Massive Agent-Based Simulations
Antelmi A.;Cordasco G.;D'Auria M.;De Vinco D.;Negro A.;Spagnuolo C.
2019-01-01
Abstract
The analysis of real systems and the development of predictive models to describe the evolution of real phenomena are challenging tasks that can improve the design of methodologies in many research fields. In this context, Agent-Based Model (ABM) can be seen as an innovative tool for modelling real-world complex simulations. This paper presents Rust-AB, an open-source library for developing ABM simulation on sequential and/or parallel computing platforms, exploiting Rust as programming language. The Rust-AB architecture as well as an investigation on the ability of Rust to develop ABM simulations are discussed. An ABM simulation written in Rust-AB, and a performance comparison against the well-adopted Java ABM toolkit MASON is also presented.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
