Metraipolis › archive › dc_9bb382b0a4e2
Argus: Orchestrating Cross-Layer GPU Performance Measurements around Semantic Regions — arXiv:2609.12299
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.12299
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T03:08:35.139Z
- SHA-256 of the archived text
85b9d19021d691c2201e6318bc4d47881e3377346bf1573dab66c0f173d0bda7- Stable URL of this page
https://metraipolis.com/documento/dc_9bb382b0a4e2
How to cite
Argus: Orchestrating Cross-Layer GPU Performance Measurements around Semantic Regions — arXiv:2609.12299 — arXiv (en), https://arxiv.org/abs/2609.12299 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9bb382b0a4e2 (Metraipolis archive dc_9bb382b0a4e2, sha256 85b9d19021d691c2201e6318bc4d47881e3377346bf1573dab66c0f173d0bda7)
BibTeX
@misc{dc9bb382b0a4e2,
title = {Argus: Orchestrating Cross-Layer GPU Performance Measurements around Semantic Regions — arXiv:2609.12299},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.12299},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9bb382b0a4e2; sha256 85b9d19021d691c2201e6318bc4d47881e3377346bf1573dab66c0f173d0bda7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.12299. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
GPU developers and automated optimizers need performance evidence for semantic code regions--such as neural-network operator implementations and pipeline stages--but this evidence is fragmented across profiling tools. Answering a region-level question can require manually constructing probes and program variants, isolating interfering measurements, and mapping evidence to regions and execution contexts. We present Argus, a region-centric measurement planner and runtime that automates this workflow. Clients identify regions with boundary markers and select signals and execution scopes. Argus preserves region identity across compilation, execution, and measurement variants, constructs interference-aware multi-run plans, and orchestrates transformations and profiling across backends. It joins compiler-, hardware-, and system-level evidence using region identity and dynamic execution context, producing reports that record measurement origins and attribution ambiguity. We evaluate Argus across agentic kernel optimization, persistent megakernel optimization, and cross-level PGO. Across 44 persistent-GEMM and attention configurations, Argus improves 39/44 cases and raises AlphaEvolve's geometric-mean speedup from 5.4% to 8.9%. On a persistent TinyLlama-1.1B decode megakernel, an optimization agent reaches 1.65 ms/token with Argus versus 4.92 ms/token without it, producing a kernel $2.1\times$ faster than PyTorch with CUDA Graphs. Finally, Argus-guided cross-level PGO improves compute--communication overlap, increasing throughput by 7% on average across five multi-GPU settings. Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-14 in cs.DC). Categorie arXiv: Distributed, Parallel, and Cluster Computing (cs.DC) ; Performance (cs.PF). Pagina del lavoro: https://arxiv.org/abs/2609.12299 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-16T03:03:05.079Z dalla pagina https://arxiv.org/catchup/cs.DC/2026-09-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).