Migrating Molecular Docking to oneAPI: the AutoDock-GPU Case Study
In this talk, we present our experiences migrating AutoDock-GPU, a widely-used molecular docking application to oneAPI. We discuss our experiences migrating from CUDA to SYCL by leveraging the Intel® DPC++ Compatibility Tool released as part of oneAPI. Our focus is on 1) the usage of the automated source-code migration tool, 2) the manual code refinements for functional correctness and performance optimization, and 3) the comparison of the performance achieved on high-end GPUs, including the recently launched Intel® Data Center Max 1550 device.
Speaker(s)
Mr. Leonardo Solis-Vasquez - Post-doctoral researcher in the Embedded Systems and Applications Group at the Technical University of Darmstadt- EMEA
Leonardo Solis-Vasquez works as a post-doctoral researcher in the Embedded Systems and Applications Group at the Technical University of Darmstadt. His research interests are high-performance and scientific computing, as well as parallel programming. Since the establishment of the oneAPI Center of Excellence at TU Darmstadt, he has been focused on accelerating parallel computing and simulation software used in medical and pharmaceutical research, powered by oneAPI and the recently-launched Intel Max Series CPUs and GPUs.
