Domain-Specific Translator and Optimizer for Massive On-Chip Parallelism
Didem Unat, Ph. D. Dissertation, Department of Computer Science and Engineering, University of California, San Diego, 2012.
Accelerating a 3D Finite-Difference Earthquake Simulation with a C-to-CUDA Translator
D. Unat, J. Zhou, Y. Cui, S. B. Baden, X. Cai, Computing in Science & Engineering, pp. 48-59, May/June 2012
Mint: Realizing CUDA performance in 3D Stencil Methods with Annotated C
D.Unat, X.Cai, and S. Baden., International Conference on Supercomputing. ICS'11, Tucson, AZ, 2011. Paper.pdf
Auto-optimization of a Feature Selection Algorithm
D.Unat, H.S. Kim, J. Schulze, S.B. Baden. in Emerging Applications and Many-core Architecture, EAMA Workshop co-located with ISCA, San Jose, CA, 2011.
"Mint Programming Model for non-expert CUDA programmers"
D.Unat and S. Baden, UCSD Research Expo, April, 2011. Poster.pdf
"Cuda-free programming of 3D stencil methods with Mint"
D. Unat, X. Cai, and S. Baden., in Manycore and Accelerator-based High Performance Scientific Computing, ICCS Workshop, Berkeley, CA, 2011. Talk.pdf
Mint: An OpenMP to CUDA Translator
D. Unat., Poster presented at the GPU Technology Conference, San Jose, CA, September 2010. Poster.pdf