SIML: A Fast SIMD Algorithm for Calculating LINGO Chemical Similarities on GPUs and CPUs
Department of Computer Science, Stanford University, Stanford, California; Department of Chemistry, Stanford University, Stanford, California
Journal of Chemical Information and Modeling, Vol. 50, No. 4. (26 April 2010), pp. 560-564.
@article{haque2010siml,
title={SIML: A Fast SIMD Algorithm for Calculating LINGO Chemical Similarities on GPUs and CPUs},
author={Haque, I.S. and Pande, V.S. and Walters, W.P.},
journal={Journal of chemical information and modeling},
volume={50},
number={4},
pages={560–564},
issn={1549-9596},
year={2010},
publisher={ACS Publications}
}
LINGOs are a holographic measure of chemical similarity based on text comparison of SMILES strings. We present a new algorithm for calculating LINGO similarities amenable to parallelization on SIMD architectures (such as GPUs and vector units of modern CPUs). We show that it is nearly 3x as fast as existing algorithms on a CPU, and over 80x faster than existing methods when run on a GPU.
November 9, 2010 by hgpu
