Preprint
Aug 2026
Architecting the Next Generation of Asynchronous, Distributed GPUs for the AI Era
This work presents a cycle-level simulation framework designed to accurately model modern GPU generations, including Ampere, Hopper, and Blackwell, and conducts architectural case studies to evaluate emerging design trajectories, including chiplet topology scaling, expanded SRAM capacity and bandwidth, and inter-GPU pr...
Junrui Pan, Wei-Li An, Cesar Avalos Baddouh et al.
· 2 citations
· ⚡2