Spiking Neural Network (SNN) simulation serves as a crucial tool for understanding brain dynamics and advancing neuromorphic computing, but faces significant scalability challenges in large-scale distributed implementations. The primary bottlenecks arise from frequent synchronization overhead in time-driven simulators...
Guang-Nan Feng, Yang-Le Zeng, Hong Liang et al.· 0 citations
CerebroSim, a scalable framework for whole-brain simulation that combines Delay-aware Spike Broadcast for aggregated delay-aware communication, Race-free Synaptic Dynamics Computation for lock/atomic-free multithreaded delivery with HBM-aware optimization, and Sparse Synapse Storage Compression for compact indexing wit...
Guang-Nan Feng, Tianxiang Lyu, Hao Huang et al.· 0 citations
Kohn-Sham density functional theory (DFT) remains the workhorse of ab initio materials simulation, yet cubic computational and quadratic memory scaling have confined calculations to a few hundred to thousands of atoms, spanning only nanometers, far below experimentally relevant length scales. We introduce XLSDFT, a lin...
Qi-Men Xu, Yu Zhang, Di-Xing Ni et al.· arXiv.org· 0 citations
LineShine is presented, a newly deployed exascale supercomputer whose online acceleration architecture provides a new design point for converged HPC and AI and demonstrates that its balanced architecture and software stack can translate system capability into high throughput, robust scalability, and sustained scientifi...