Hello.
I'm a PhD candidate in The
Department of Computer Science at The University of Oregon.
I study the design of parallel algorithms, with particular
care towards numerical methods for large-scale scientific
computing.
In such domains, computational constraints determine the scale
and cost of scientific simulation;
as hardware has approached physical limits,
sustained performance increasingly requires treating algorithms
and hardware as a single, coherent design problem.
My work seeks to both optimize existing applications and to
find novel algorithms that more effectively utilize modern
hardware, including GPU accelerators and emerging
high-performance architectures.
EMAIL — josephmcl « at » protonmail « dot » ch
Technical Writing
- McLaughlin J.., Choi J., ORC-IR: Achieving Near Double Precision Accuracy of Dense Linear Solves in Half Precision. In Review.
- McLaughlin J. The Role of Matrix Structure on PDE Solver Performance. University of Oregon. Area Exam. 2025 December 4. PDF TALK
- McLaughlin J., Erickson B., Choi J., An Algorithm for Fast Hybridized PDE Solvers. In Review.
- McLaughlin J., Choi J., Durairajan R., ×Grid: A Location-oriented Topology Design for LEO Satellites. LEO-NET 2023—1st International Workshop on LEO Networking and Communication at ACM MobiCom 2023, Madrid, Spain, 2023 October 2—3. PDF DOI
- McLaughlin J., A high-performance profile of hybridized PDEs. University of Oregon. 2023 November 28. PDF TALK
- McLaughlin J., Scoping study: high-efficiency X-ray sources for STARBRIGHT. Lawrence Livermore National Laboratory (LLNL) internal audience, publicized. 2023 September 12. PDF
- McLaughlin J., Young M., Rockcastle S., Fine-grained measurement of the indoor built environment with robotic vacuum cleaners. BS2021—17th International Conference of the International Building Performance Simulation Association, Bruges, 2021 September 1–3. PDF DOI
- Flores J., Fuentes R., McLaughlin J., Novitzky K., Schofield S., Springel A., Tal S., Pipeline Trees - An Auxiliary Tool in the Creation of Time Series Pipelines, ITISE2021—International Conference on Time Series and Forecasting, Gran Canaria, 2021 July 19–21. PDF