作者位次 #11/15OSDI 2026Oral· Operating Systems & Distributed Systems
MoonBright: A GPU Memory Allocator with Device-Side Page Table Materialization and Deferred TLB Coherence
Yangyu Zhang · Lei Chen · Chunwei Xia · Shuaijiang Li · Shuoming Zhang · Zhicheng Li · Qianqi Sun · Jiawei Xiao · Ruiyuan Xu · Ao Chen · Guangli Li · Xiaobing Feng · Huimin Cui · Chenxi Wang · Jiacheng Zhao
摘要
Modern GPU workloads increasingly rely on dynamic and fine-grained memory allocation, yet GPU memory management remains CPU-centric. In current GPU runtimes, allocation metadata updates, page-table construction, translation-state propagation, and TLB shootdowns are largely serialized through the...