Research

I am a part of the HPC-lab at the Department of Computer and Information Science.
My advisor is Dr. Anne C. Elster.
My Google Scholar profile

Publications

Thomas L. Falch, Anne C. Elster
Machine Learning Based Auto-Tuning for Enhanced Performance Portability of OpenCL Applications.
Concurrency and Computation: Practice and Experience, 2016
Available here Code and aditional results available here

Thomas L. Falch, Anne C. Elster
ImageCL: An image processing language for performance portability on heterogeneous systems.
In proc. 2016 International Conference on High Performance Computing and Simulation (HPCS)
Available here A PDF for personal use, which also includes aditional results, is available here Code available here

Nico Reissmann, Thomas L. Falch, Benjamin A. Bjornseth, Helge Bahmann, Jan Christian Meyer, Magnus Jahre
Efficient control flow restructuring for GPUs
In proc. 2016 International Conference on High Performance Computing and Simulation (HPCS)
Available here

Thomas L. Falch, Anne C. Elster
Machine Learning Based Auto-Tuning for Enhanced OpenCL Performance Portability
In proc. 2015 IEEE International Parallel and Distributed Processing Symposium Workshop (IPDPSW)
Available here PDF for personal use available here Code available here

Thomas L. Falch, Anne C. Elster
Register Caching for Stencil Computations on GPUs
In proc. 16th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC '14)
Available here Code available here

Erik Smistad, Thomas L. Falch, Mohammadmehdi Bozorgi, Anne C. Elster, Frank Lindseth
Medical Image Segmentation on GPUs - A Comprehensive Review
Medical Image Analysis
Available here

Thomas L. Falch, Jostein B. Fl√łystad,Dag W. Breiby, Anne C. Elster
GPU-Accelerated Visualization of Scattered Point Data
IEEE Access
Available here Code available here

Thomas L. Falch, Jostein B. Fl√łystad, Anne C. Elster, Dag W. Breiby
Optimization and Parallelization of Ptychography Reconstruction code
In proc. 25th Norwegian Informatics Conference (NIK'12)
Available here