Metraipolis › archive › dc_74a21293668f
Characterizing Contention-Induced Reliability Collapse in KV-Cache Timing Side Channels for Multi-Tenant LLM Serving — arXiv:2609.06853
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.06853
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T13:40:15.859Z
- SHA-256 of the archived text
8c6c89139469117fdd881245f0f665841eb5a5ee94fd72654d4b6916ad9c3c1c- Stable URL of this page
https://metraipolis.com/documento/dc_74a21293668f
How to cite
Characterizing Contention-Induced Reliability Collapse in KV-Cache Timing Side Channels for Multi-Tenant LLM Serving — arXiv:2609.06853 — arXiv (en), https://arxiv.org/abs/2609.06853 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_74a21293668f (Metraipolis archive dc_74a21293668f, sha256 8c6c89139469117fdd881245f0f665841eb5a5ee94fd72654d4b6916ad9c3c1c)
BibTeX
@misc{dc74a21293668f,
title = {Characterizing Contention-Induced Reliability Collapse in KV-Cache Timing Side Channels for Multi-Tenant LLM Serving — arXiv:2609.06853},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.06853},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_74a21293668f; sha256 8c6c89139469117fdd881245f0f665841eb5a5ee94fd72654d4b6916ad9c3c1c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06853. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Shared key--value (KV) cache reuse improves large language model (LLM) serving, but it can also create a timing side channel that reveals whether a prefix is already cached. Previous work shows that such attacks are possible, but their reliability under realistic multi-tenant contention is less understood. We study this problem through seven experiments on live shared LLM-serving systems. On a vLLM server running DeepSeek-R1-Distill-Llama-8B on NVIDIA GB10, mean Cohen's d drops from 0.7789 with no synthetic workers to 0.2109 with two workers (t=8.412), while higher worker counts cause no statistically detectable further loss. A 120-run sparse-overlap experiment places the best breakpoint at the boundary of the measured range (tau=0, 95% CI [0.000,0.113]), supporting an ambient-versus-loaded regime change rather than an internal physical threshold. AUROC falls from 0.650 at ambient to 0.531 near 61% overlap and partially recovers to 0.574 at saturation. Concurrency-depth variance is the strongest measured correlate of effect size (r=-0.416) and hit consistency (r=-0.637). An interleaved control preserves the same non-monotonic ordering. The main collapse is also reproduced on a real two-node, two-GPU tensor-parallel vLLM setup, where mean d falls from 3.418 to 0.511 (p<0.01). Two SGLang pilots are statistically inconclusive. Overall, KV-cache timing reliability depends strongly on the load regime and serving stack, and measurements on quiet systems can overestimate operational attack reliability. Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CR). Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.06853 — 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 2026-10-16T13:09:48.446Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).