Parallelisation of Java for Graphics Processors
Trinity College
University of Cambridge Computer Laboratory, 2010
@article{calvert2010parallelisation,
title={Parallelisation of Java for Graphics Processors},
author={Calvert, P.},
journal={Final-year dissertation at University of Cambridge Computer Laboratory. Available from http://www. cl. cam. ac. uk/~ prc33},
year={2010}
}
The aim of the project was to allow extraction and compilation of Java virtual machine bytecode for parallel execution on graphics cards, specifically the NVIDIA CUDA framework, by both explicit and automatic means. The compiler, which was produced, successfully extracts and compiles code from class files into CUDA C++ code, and outputs transformed classes that make use of this native code. Developers can indicate loops that should be parallelised by use of Java annotations. Loops can also be automatically detected as ‘safe’ using a dependency checking algorithm. On benchmarks, speedups of up to a factor of 187 were measured. Evaluation of the automatic dependency analysis showed 85% accuracy over a range of sample code.
October 6, 2011 by hgpu
