Barracuda: a dynamic, Turing-complete GPU virtual machine for high-performance simulations
Medical & Biological Engineering & Computing · 2026
DOI: 10.1007/s11517-025-03438-3 ↗Kurz gefasst
Fachartikel in der Zeitschrift Medical & Biological Engineering & Computing, 2026. Verfasst von Phillip Duncan-Gelder und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Physical and Chemical Sciences, University of Canterbury, Christchurch, New Zealand. phillip.duncan-gelder@southerndhb.govt.nz. Hauptthemen laut MEDLINE: Computer Simulation, Computer Graphics, Software, User-Computer Interface. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Canterbury.
Abstract (Originalsprache)
Accurate simulation of dynamic biological phenomena, such as tissue response and disease progression, is crucial in biomedical research and diagnostics. Traditional GPU-based simulation frameworks, typically static CUDA ® environments, struggle with dynamically evolving parameters, limiting flexibility and clinical applicability. We introduce Barracuda, an open-source, lightweight, header-only, Turing-complete virtual machine designed for seamless integration into GPU environments. Barracuda enables real-time parameter perturbations through an expressive instruction set and operations library, implemented in a compact C/CUDA library. A dedicated high-level programming language and Rust-based compiler enhance accessibility, allowing straightforward integration into biomedical simulation workflows. Benchmark validations, including Rule 110 cellular automaton and Mandelbrot computations, confirm Barracuda’s versatility and computational completeness. In magnetic resonance imaging (MRI) simulations, Barracuda allows for the dynamic recalculation of critical parameters, such as $$T_1$$ T 1 relaxation times and temperature-induced off-resonance frequencies. Although it introduces computational overhead compared to static kernels, Barracuda significantly improves simulation accuracy by enabling dynamic modeling of key biological processes. Barracuda’s modular architecture supports incremental integration, providing valuable flexibility for biomedical research and rapid prototyping. Future developments aim to optimize performance and expand domain-specific instruction sets, reinforcing Barracuda’s role in bridging static GPU programming and dynamic simulation requirements. Graphic abstract
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Physical and Chemical Sciences, University of Canterbury, Christchurch, New Zealand. phillip.duncan-gelder@southerndhb.govt.nz.
- Te Whatu Ora - Health New Zealand, New Zealand, . phillip.duncan-gelder@southerndhb.govt.nz.
- School of Physical and Chemical Sciences, University of Canterbury, Christchurch, New Zealand.
- Te Whatu Ora - Health New Zealand, New Zealand.
- Electrical and Computer Engineering, University of Canterbury, Christchurch, New Zealand.
ORCID-Profile: Phillip Duncan-Gelder, Darin O'Keeffe, Steven Marsh
Förderung
- University of Canterbury
Zitieren
APA
Duncan-Gelder, P., O'Keeffe, D., Bones, P. J., & Marsh, S. (2026). Barracuda: a dynamic, Turing-complete GPU virtual machine for high-performance simulations. Medical & Biological Engineering & Computing. https://doi.org/10.1007/s11517-025-03438-3
Vancouver
Duncan-Gelder P, O'Keeffe D, Bones PJ, Marsh S. Barracuda: a dynamic, Turing-complete GPU virtual machine for high-performance simulations. Medical & Biological Engineering & Computing. 2026. doi:10.1007/s11517-025-03438-3
BibTeX
@article{DuncanGelder2026,
author = {Duncan-Gelder, Phillip and O'Keeffe, Darin and Bones, Philip J and Marsh, Steven},
title = {Barracuda: a dynamic, Turing-complete GPU virtual machine for high-performance simulations},
journal = {Medical & Biological Engineering & Computing},
year = {2026},
doi = {10.1007/s11517-025-03438-3}
}
EUVIMED