Accelerating molecular docking leveraging DPC++ and Intel CPUs and GPUs
- 0 Collaborators
This project aims to port an accelerated version of the AutoDock molecular docking application to create a single codebase that can be efficiently optimized and tuned for multiple hardware architecture targets. ...learn more
Project status: Under Development
Intel Technologies
DevCloud,
oneAPI,
Migrated To SYCL,
Intel vTune,
Intel CPU,
DPC++
Overview / Usage
Molecular docking simulations aim to predict the best fit between two molecules. These “docking” results are an important initial step for the discovery of new drugs, as the computations can be performed much more quickly than experiments using traditional “wet lab” chemistry. Such simulations involve compute-intensive computations, and thus, can profit a lot from hardware acceleration. In this project, we focus on AutoDock-GPU, an OpenCL & CUDA implementation of the Scripps' popular AutoDock molecular docking tool. Concretely, we aim to aim to develop a DPC++ version of AutoDock-GPU to target next-generation of Intel CPUs and GPUs.
Methodology / Approach
We plan to port current OpenCL/CUDA implementation into a DPC++ one. For that purpose, we will use oneAPI tools, such as Intel Advisor and Intel VTune Profiler.
Technologies Used
oneAPI Base Toolkit, Data-Parallel C++, Intel Integrated Graphics Gen 9.5,