计算机科学
并行计算
GPU群集
图形处理单元
加速度
计算科学
库达
韦尔莱积分法
图形处理单元的通用计算
加速
绘图
中央处理器
分子动力学
计算机硬件
计算机图形学(图像)
化学
物理
计算化学
经典力学
作者
Szilárd Páll,Artem Zhmurov,Paul Bauer,M Abraham,Magnus Lundborg,Alan Gray,Berk Hess,Erik Lindahl
摘要
The introduction of accelerator devices such as graphics processing units (GPUs) has had profound impact on molecular dynamics simulations and has enabled order-of-magnitude performance advances using commodity hardware. To fully reap these benefits, it has been necessary to reformulate some of the most fundamental algorithms, including the Verlet list, pair searching and cut-offs. Here, we present the heterogeneous parallelization and acceleration design of molecular dynamics implemented in the GROMACS codebase over the last decade. The setup involves a general cluster-based approach to pair lists and non-bonded pair interactions that utilizes both GPUs and CPU SIMD acceleration efficiently, including the ability to load-balance tasks between CPUs and GPUs. The algorithm work efficiency is tuned for each type of hardware, and to use accelerators more efficiently we introduce dual pair lists with rolling pruning updates. Combined with new direct GPU-GPU communication as well as GPU integration, this enables excellent performance from single GPU simulations through strong scaling across multiple GPUs and efficient multi-node parallelization.
科研通智能强力驱动
Strongly Powered by AbleSci AI