Reproduce “TailGuard-GDS: Tail-Safe Control of GPU Storage Data Movement” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jiaqi Liu, Fan Yang, Xiaohang Shi, Xianzhe Zhang, Tao Jiang, Guojun Yuan, Zhan Wang, Guangming Tan
TailGuard-GDS selects among equivalent storage-to-GPU-buffer delivery plans to reduce delay to dependent GPU work under phase throughput or deadline requirements. It measures plans across backend, request shape, and eligible placement choices. In matched-load tests, host staging had lower p999 in 84 of 100 cases. Across five eight-GPU ZeRO-3 workloads, it reduced step p95 by 7.9–24.6% on the authors’ eight-H100 system.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.