OPAC
Optimization of Pattern matching Compression Algorithms for GPUs
Optimization of Pattern matching Compression Algorithms for GPUs
The main goal of the project was to study the usage of parallel hardware platforms and software stacks (CUDA, OpenCL) to speed up the execution of recurrent pattern matching-based compression algorithms. The hardware platforms include off the shelves desktop machines, multi-core CPUs, System on a Chip (SoC) and many-core Graphic Processing Units (GPUs). The project started in 2014.09.01 and ended in 2017.05.31.
Instituto de Telecomunicações / Delegação Coimbra / Portugal
Leiria (MSP-Lr) and Coimbra (IT-Co)
"Energy-Efficient and Portable Least Squares Prediction for Image Coding on a Mobile GPU" , P. Cordeiro, G. Falcao, P. Domingues, N. Rodrigues and S. Faria, 25th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP), St. Petersburg, 2017, pp. 237-240. https://doi.org/10.1109/PDP.2017.97
"Optimized fast Walsh-Hadamard transform on OpenCL-GPU and OpenCL-CPU," P. M. M. Pereira, P. Domingues, N. M. M. Rodrigues, S. M. M. Faria and G. Falcao, Sixth International Conference on Image Processing Theory, Tools and Applications (IPTA), Oulu, 2016, pp. 1-6. https://doi.org/10.1109/IPTA.2016.7820984
“Optimizing GPU Code for CPU Execution Using OpenCL and Vectorization: A Case Study on Image Coding”, PMM Pereira, P Domingues, NMM Rodrigues, G Falcao, SMM de Faria, International Conference on Algorithms and Architectures for Parallel Processing - ICA3PP 2016: Algorithms and Architectures for Parallel Processing pp 537-545, Granada, Spain, 2016. https://link.springer.com/chapter/10.1007/978-3-319-49583-5_42
“Assessing the performance and energy usage of multi-CPUs, multi-core and many-core systems: the MMP image encoder case study”, PMM Pereira, P Domingues, NMM Rodrigues, G Falcao, SMM Faria - International Journal of Distributed and Parallel Systems (IJDPS) Vol.7, No.5, September 2016 (http://aircconline.com/ijdps/V7N5/7516ijdps01.pdf)
"OpenCL-MMP – Codificação de Imagens com Sistemas com Múltiplos Núcleos", João Filipe Crespo Silva, MSc thesis, MSc Thesis, ESTG – Polytechnic Institute of Leiria, May 2015. (http://hdl.handle.net/10400.8/1383)
“Efficient and Portable Least Squares Prediction for Image Coding on Manycore Platforms”, Pedro Nuno Henriques Maia Cordeiro, MSc Thesis, DEE/FCT/University of Coimbra, Portugal, September 2015. (http://hdl.handle.net/10316/40539)
“Optimization of Pattern Matching Algorithms for Multi- and Many-Core Platforms”, Pedro Miguel Pereira Marques, MSc Thesis, ESTG – Polytechnic Institute of Leiria, September 2016. (http://hdl.handle.net/10400.8/2267)
Source code for the OpenCL-based Walsh-Hadmard is available under an open source license at the following github address: https://github.com/ITLeiria/FWHT-OpenCL.