arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.13140cs.LGcs.CE
收藏

FlashSchNet: 快速且准确的粗粒化神经网络分子动力学

FlashSchNet: Fast and Accurate Coarse-Grained Neural Network Molecular Dynamics

Pingzhi Li, Hongxuan Li, Zirui Liu, Xingcheng Lin, Tianlong Chen

AI总结:

FlashSchNet通过I/O优化和高效算法实现快速准确的粗粒化分子动力学模拟,超越经典力场并保持高精度。

摘要:

图神经网络(GNN)势能如SchNet通过学习多体相互作用提高了分子动力学(MD)模拟的精度和可迁移性,但仍然比经典力场慢,因为碎片化的内核和内存受限的流水线未能充分利用GPU。我们展示出一个缺失的原则是使GNN-MD具备I/O意识,仔细考虑GPU高速缓存(HBM)与芯片上的SRAM之间的读写。我们提出了FlashSchNet,一个高效且准确的I/O意识SchNet风格的GNN-MD框架,基于四种技术:(1)闪速径向基函数,将成对距离计算、高斯基函数展开和余弦包络融合为单次 tiled 传递,计算每个距离一次并在所有基函数中重复使用;(2)闪速消息传递,将截断、邻居收集、滤波乘法和减少融合,以避免在HBM中材料化边张量;(3)闪速聚合,通过CSR段减少重新公式化散射-加法,将原子写入减少因素为特征维度,使正向和反向传递中的累积无竞争;(4)通道级16位量化,利用SchNet MLP权重中的低通道动态范围进一步提高吞吐量,同时几乎不损失精度。在单个NVIDIA RTX PRO 6000上,FlashSchNet在64个并行副本上实现了1000 ns/天的总模拟吞吐量,针对粗粒化(CG)蛋白质(含269个珠子)的模拟速度比CGSchNet基线快6.5倍,内存峰值减少80%,超越经典力场(如MARTINI),同时保留SchNet级别的精度和可迁移性。

英文摘要:

Graph neural network (GNN) potentials such as SchNet improve the accuracy and transferability of molecular dynamics (MD) simulation by learning many-body interactions, but remain slower than classical force fields due to fragmented kernels and memory-bound pipelines that underutilize GPUs. We show that a missing principle is making GNN-MD IO-aware, carefully accounting for reads and writes between GPU high-bandwidth memory (HBM) and on-chip SRAM. We present FlashSchNet, an efficient and accurate IO-aware SchNet-style GNN-MD framework built on four techniques: (1) flash radial basis, which fuses pairwise distance computation, Gaussian basis expansion, and cosine envelope into a single tiled pass, computing each distance once and reusing it across all basis functions; (2) flash message passing, which fuses cutoff, neighbor gather, filter multiplication, and reduction to avoid materializing edge tensors in HBM; (3) flash aggregation, which reformulates scatter-add via CSR segment reduce, reducing atomic writes by a factor of feature dimension and enabling contention-free accumulation in both forward and backward passes; (4) channel-wise 16-bit quantization that exploits the low per-channel dynamic range in SchNet MLP weights to further improve throughput with negligible accuracy loss. On a single NVIDIA RTX PRO 6000, FlashSchNet achieves 1000 ns/day aggregate simulation throughput over 64 parallel replicas on coarse-grained (CG) protein containing 269 beads (6.5x faster than CGSchNet baseline with 80% reduction of peak memory), surpassing classical force fields (e.g. MARTINI) while retaining SchNet-level accuracy and transferability.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑