Metraipolis › archive › dc_161557d486fe
PASCAL: A Progress Divergence-Aware Shared-Cache Model — arXiv:2609.10515
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.10515
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T07:54:05.187Z
- SHA-256 of the archived text
e14e874a42f9036171fc5d23d74cdc61e0b9d42fa30cd040f7f606b87e16251e- Stable URL of this page
https://metraipolis.com/documento/dc_161557d486fe
How to cite
PASCAL: A Progress Divergence-Aware Shared-Cache Model — arXiv:2609.10515 — arXiv (en), https://arxiv.org/abs/2609.10515 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_161557d486fe (Metraipolis archive dc_161557d486fe, sha256 e14e874a42f9036171fc5d23d74cdc61e0b9d42fa30cd040f7f606b87e16251e)
BibTeX
@misc{dc161557d486fe,
title = {PASCAL: A Progress Divergence-Aware Shared-Cache Model — arXiv:2609.10515},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.10515},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_161557d486fe; sha256 e14e874a42f9036171fc5d23d74cdc61e0b9d42fa30cd040f7f606b87e16251e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.10515. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In modern AI accelerators and GPUs, many concurrent cores repeatedly access the same shared data. This pattern occurs in attention, where different query (Q) tiles share the same key and value (K/V) blocks, GEMM, where every tile in a row reads the same slice, and many other operators. We name this pattern shared cyclic scan. Due to a significant amount of data reuse in this pattern, the cache is expected to capture as much data reuse as possible and largely reduce requests sent to the main memory for both performance and energy consumption concerns. However, in reality, because of the intrinsic asynchrony of multi-cores, the actual cache miss rate and DRAM traffic can be much higher compared to ideal cases. In this paper, we propose PASCAL, a shared-cache model for shared cyclic scans. It calibrates finite-run traffic, which reflects progress divergence, on reference configurations and interpolates it along static program structure such as occupancy to predict the miss rate before execution. PASCAL supports software configuration exploration without target traces or counters at scales where cycle-accurate simulation is impractical, while its analysis gives a sharp $2\sigma-1$ sufficient capacity condition for preserving LRU sharing. Across held-out scans and GEMM on GB10 and Thor, PASCAL reaches 12.04% balanced fill-equivalent miss-rate MAPE. Replacing TileSight's cache component with PASCAL lowers GB10 GEMM latency MAPE from 18.17% to 13.14%. Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-10 in cs.DC). Categorie arXiv: Performance (cs.PF) ; Hardware Architecture (cs.AR); Distributed, Parallel, and Cluster Computing (cs.DC). Pagina del lavoro: https://arxiv.org/abs/2609.10515 — 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 2026-10-16T07:25:19.026Z dalla pagina https://arxiv.org/catchup/cs.DC/2026-09-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).