GPU Acceleration of FSM Input Execution: Artifacts

  • Vanya Yaneva-Cormack (Creator)



In model-based development, software is implemented and verified based on a model of the required system. Finite State Machines (FSMs) are widely used as models in several domains but validating that they accurately represent the required behaviour requires the execution of many input sequences, which is often expensive and time-consuming. We research the use of Graphics Processing Units (GPUs) to accelerate FSM input execution for the purposes of FSM model validation. This dataset contains the experimental data and scripts associated with our work on improving the performance and scalability of this approach, which is published in the below paper. The dataset contains the following:
1. subject FSMs used in experimentation,
2. a C program for the generation of input sequences input sequences based on the all-transition-pair coverage criterion,
3. an implementation of the input reduction algorithm used in the paper,
4. experimental data for input execution time on the GPU and a 16-core CPU,
5. R scripts for result analysis and plot generation, used in the paper

Data Citation

Yaneva-Cormack, Vanya. (2021). GPU Acceleration of FSM Input Execution: Artifacts, [software]. University of Edinburgh. School of Informatics.
Date made available1 Oct 2021
PublisherEdinburgh DataShare

Cite this