Other formats:
BibTeX
LaTeX
RIS
@article{1505476, author = {Střelák, David and Sorzano, Carlos Óscar S. and Carazo, José María and Filipovič, Jiří}, article_number = {5}, doi = {http://dx.doi.org/10.1177/1094342019832958}, keywords = {Cryo-EM; GPU; CUDA; 3-D Fourier reconstruction; auto-tuning}, language = {eng}, issn = {1094-3420}, journal = {The International Journal of High Performance Computing Applications}, title = {A GPU acceleration of 3-D Fourier reconstruction in cryo-EM}, url = {https://journals.sagepub.com/doi/10.1177/1094342019832958}, volume = {33}, year = {2019} }
TY - JOUR ID - 1505476 AU - Střelák, David - Sorzano, Carlos Óscar S. - Carazo, José María - Filipovič, Jiří PY - 2019 TI - A GPU acceleration of 3-D Fourier reconstruction in cryo-EM JF - The International Journal of High Performance Computing Applications VL - 33 IS - 5 SP - 948-959 EP - 948-959 PB - SAGE Publishing SN - 10943420 KW - Cryo-EM KW - GPU KW - CUDA KW - 3-D Fourier reconstruction KW - auto-tuning UR - https://journals.sagepub.com/doi/10.1177/1094342019832958 L2 - https://journals.sagepub.com/doi/10.1177/1094342019832958 N2 - Cryo-electron microscopy is a popular method for macromolecules structure determination. Reconstruction of a 3-D volume from raw data obtained from a microscope is highly computationally demanding. Thus, acceleration of the reconstruction has a great practical value. In this article, we introduce a novel graphics processing unit (GPU)-friendly algorithm for direct Fourier reconstruction, one of the main computational bottlenecks in the 3-D volume reconstruction pipeline for some experimental cases (particularly those with a large number of images and a high internal symmetry). Contrary to the state of the art, our algorithm uses a gather memory pattern, improving cache locality and removing race conditions in parallel writing into the 3-D volume. We also introduce a finely tuned CUDA implementation of our algorithm, using auto-tuning to search for a combination of optimization parameters maximizing performance on a given GPU architecture. Our CUDA implementation is integrated in widely used software Xmipp, version 3.19, reaching 11.4× speedup compared to the original parallel CPU implementation using GPU with comparable power consumption. Moreover, we have reached 31.7× speedup using four GPUs and 2.14×–5.96× speedup compared to optimized GPU implementation based on a scatter memory pattern. ER -
STŘELÁK, David, Carlos Óscar S. SORZANO, José María CARAZO and Jiří FILIPOVIČ. A GPU acceleration of 3-D Fourier reconstruction in cryo-EM. \textit{The International Journal of High Performance Computing Applications}. SAGE Publishing, 2019, vol.~33, No~5, p.~948-959. ISSN~1094-3420. Available from: https://dx.doi.org/10.1177/1094342019832958.
|