Analytical Provisioning for Attention-FFN Disaggregated LLM Serving under Stochastic Workloads

[2601.21351] Analytical Provisioning for Attention-FFN Disaggregated LLM Serving under Stochastic Workloads Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Machine Learning arXiv:2601.21351 (cs) [Submitted on 29 Jan 2026 ( v1 ), last revised 14 Aug 2026 (this version, v4)] Title: Analytical Provisioning for Attention-FFN Disaggregated LLM Serving under Stochastic Workloads Authors: Chendong Song , Meixuan Wang , Hang Zhou , Hong Liang , Yuan Lyu , Zixi Chen , Yuwei Fan , Zijie Zhou View a PDF of the paper titled Analytical Provisioning for Attention-FFN Disaggregated LLM Serving under Stochastic Workloads, by Chendong Song and 7 other authors View PDF HTML (experimental) Abstract: Attentio-FFN disaggregation (AFD) is an emerging architecture for LLM decoding that separates state-heavy, KV-cache-dominated Attention computation from stateless, compute-intensive FFN computation, connected by per-step communication. While AFD enables independent scaling of memory and compute resources, its performance is highly sensitive to the Attention/FFN provisioning ratio: mis-sizing induces step-level blocking and costly device idle time. We develop an analytical provisioning framework for AFD bundles in an $r$A--$1$F topology under stochastic workloads. Two sources of randomness shape the problem: per-slot Attention workload evolves as KV caches grow and completed requests are replenished with random prompt and decode lengths, and synchronized execution across Attention workers introduces a barrier governed by the slowest worker. We address both via a renewal-reward characterization of the per-slot stationary token load, identifying a single workload statistic $\theta$ that governs provisioning under arbitrary prefill-decode distributions and admits a nonparametric estimator from request traces. The analysis yields a closed-form mean-field rule for the optimal A/F ratio decomposing into Attention-, communication-, and FFN-bottleneck regimes, together with a Gaussian barrier-aware refinement that quantifies cross-worker synchronization overhead. A trace-calibrated AFD simulator supports the framework across workloads: the predicted optimal ratio matches the simulation-optimal within 10%. Together, these results provide a compact, calibratable account of how stochastic workload structure determines provisioning in disaggregated LLM serving. Comments: Submitted to Neurips 2026 Subjects: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI) Cite as: arXiv:2601.21351 [cs.LG] (or arXiv:2601.21351v4 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2601.21351 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Chendong Song [ view email ] [v1] Thu, 29 Jan 2026 07:22:27 UTC (2,185 KB) [v2] Thu, 7 May 2026 05:13:35 UTC (1,540 KB) [v3] Tue, 12 May 2026 02:35:02 UTC (1,540 KB) [v4] Fri, 14 Aug 2026 20:47:28 UTC (1,540 KB) Full-text links: Access Paper: View a PDF of the paper titled Analytical Provisioning for Attention-FFN Disaggregated LLM Serving under Stochastic Workloads, by Chendong Song and 7 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.LG < prev | next > new | recent | 2026-01 Change to browse by: cs cs.AI References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) IArxiv recommender toggle IArxiv Recommender ( What is IArxiv? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from

Related documents

• DASC: Decay-Aware State Compression for Hybrid Linear-Attention Serving#799590
• Strategy and Motivation, Rather Than Fatigue, Drive Age‐Related Differences in Sustained Attention Performance: Evidence for Decoupled Beta‐Band Oscillations#240940
• Persistent Memory Through Triple-Loop Consolidation Under Stochastic Unit Turnover#799988
• ESAA-Conversational: An Event-Sourced Memory Layer for Continuity, Handoff, and Curation Across Heterogeneous LLM Coding Agents#303298
• The Illusion of Power Capping in LLM Decode: A Phase-Aware Energy Characterisation Across Attention Architectures#178843