Optimizing Alloy Models

Abstract

This paper presents three possible optimizations of Alloy models, including how and when to implement these optimizations. Alloy is a formal light-weight language for performing incremental and automatic analysis. Analysis is performed within a user-defined scope, which limits the number of model elements that are considered. When this scope increases, the number of possible combinations of model elements increases exponentially. Thus the analysis time escalates rapidly caused by this state-space explosion. Implementing the optimizations presented in this paper will decrease the analysis time, and thus make analysis suitable for larger models. We give concrete examples showing the decrease in analyzation effort and time given these optimizations.

Oppdragsgiver: Research Council of Norway