Theoretically Optimal Attention/FFN Ratios in Disaggregated LLM Serving

• URL: http://arxiv.org/abs/2601.21351v1
• Date: Thu, 29 Jan 2026 07:22:27 GMT
• Title: Theoretically Optimal Attention/FFN Ratios in Disaggregated LLM Serving
• Abstract summary: We develop a tractable analytical framework for sizing AFD bundles in a $r$A-$1$F topology.<n>We derive closed-form rules for the optimal A/F ratio that maximize average throughput per instance.<n>A trace-calibrated AFD simulator validates the theory.
• Score: 12.150964112168582
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Attention-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 a tractable analytical framework for sizing AFD bundles in an $r$A-$1$F topology, where the key difficulty is that Attention-side work is nonstationary-token context grows and requests are continuously replenished with random lengths-while FFN work is stable given the aggregated batch. Using a probabilistic workload model, we derive closed-form rules for the optimal A/F ratio that maximize average throughput per instance across the system. A trace-calibrated AFD simulator validates the theory: across workloads, the theoretical optimal A/F ratio matches the simulation-optimal within 10%, and consistently reduces idle time.

Related papers

• Faster Than Flash: Exploiting Attention Sparsity for Efficient Long-Context Decoding [51.48425605758328]
  We present Faster Flash Decoding (FFD), a novel hardware-algorithm co-design framework designed to break the memory wall in long-context decoding.<n>FFD integrates the selector and computer into a fully fused kernel, replacing external metadata indices with content-aware scanning.<n>It achieves up to 11.6x kernel-level speedup and scaling to 256K context length, with 2.37x end-to-end throughput improvement.
  arXiv  Detail & Related papers  (2026-08-31T15:13:20Z)
• From Atoms to Entropy: Optimal Noise Allocation for Diffusion Training in the Convex Regime [52.459513910242485]
  We develop a general statistical framework for studying optimal noise-level allocation in diffusion training.<n>We show that the optimized training schedule admits an atomic minimizer, concentrated on finitely many noise levels.<n>We then evaluate the entropic schedule in larger-scale experiments, where full schedule optimization is currently intractable.
  arXiv  Detail & Related papers  (2026-07-10T20:59:06Z)
• How Far Can Disaggregation Go? A Design-Space Exploration of Attention-FFN Disaggregation for Efficient MoE LLM Serving [14.552785121277529]
  We study the benefits and limits of Attention-FFN Disaggregation (AFD) at scale through a framework that fuses on-device kernel measurements with high-fidelity network simulation.<n>Under strict TTFT/TPOT SLOs, AFD sustains around 4k tokens/s of system throughput on DeepSeek-V3.2 across chat, coding, and agentic-coding workloads.
  arXiv  Detail & Related papers  (2026-05-27T10:55:57Z)
• Fractional-Order Federated Learning [4.1751058176413105]
  Federated learning (FL) allows remote clients to train a global model collaboratively while protecting client privacy.<n>Despite its privacy-preserving benefits, FL has significant drawbacks, including slow convergence, high communication cost, and non-independent-and-identically-distributed (non-IID) data.
  arXiv  Detail & Related papers  (2026-02-17T06:25:23Z)
• Causal Autoregressive Diffusion Language Model [70.7353007255797]
  CARD reformulates the diffusion process within a strictly causal attention mask, enabling dense, per-token supervision in a single forward pass.<n>Our results demonstrate that CARD achieves ARM-level data efficiency while unlocking the latency benefits of parallel generation.
  arXiv  Detail & Related papers  (2026-01-29T17:38:29Z)
• Efficiency vs. Fidelity: A Comparative Analysis of Diffusion Probabilistic Models and Flow Matching on Low-Resource Hardware [0.0]
  Denoising Diffusion Probabilistic Models (DDPMs) have established a new state-of-the-art in generative image synthesis.<n>This study presents a comparative analysis of DDPMs against the emerging Flow Matching paradigm.
  arXiv  Detail & Related papers  (2025-11-24T18:19:42Z)
• Adaptive Deadline and Batch Layered Synchronized Federated Learning [66.93447103966439]
  Federated learning (FL) enables collaborative model training across distributed edge devices while preserving data privacy, and typically operates in a round-based synchronous manner.<n>We propose ADEL-FL, a novel framework that jointly optimize per-round deadlines and user-specific batch sizes for layer-wise aggregation.
  arXiv  Detail & Related papers  (2025-05-29T19:59:18Z)
• The Larger the Merrier? Efficient Large AI Model Inference in Wireless Edge Networks [56.37880529653111]
  The demand for large computation model (LAIM) services is driving a paradigm shift from traditional cloud-based inference to edge-based inference for low-latency, privacy-preserving applications.<n>In this paper, we investigate the LAIM-inference scheme, where a pre-trained LAIM is pruned and partitioned into on-device and on-server sub-models for deployment.
  arXiv  Detail & Related papers  (2025-05-14T08:18:55Z)
• Decentralized Nonconvex Composite Federated Learning with Gradient Tracking and Momentum [78.27945336558987]
  Decentralized server (DFL) eliminates reliance on client-client architecture.<n>Non-smooth regularization is often incorporated into machine learning tasks.<n>We propose a novel novel DNCFL algorithm to solve these problems.
  arXiv  Detail & Related papers  (2025-04-17T08:32:25Z)
• Optimizing LLM Inference: Fluid-Guided Online Scheduling with Memory Constraints [14.341123057506827]
  Large Language Models (LLMs) are indispensable in today's applications, but their inference procedure demands significant computational resources.<n>This paper formulates LLM inference optimization as a multi-stage online scheduling problem.<n>We develop a fluid dynamics approximation to provide a tractable benchmark that guides algorithm design.
  arXiv  Detail & Related papers  (2025-04-15T16:00:21Z)
• Learning the Optimal Stopping for Early Classification within Finite Horizons via Sequential Probability Ratio Test [11.199585259018459]
  Time-sensitive machine learning benefits from Sequential Probability Ratio Test (SPRT), which provides an optimal stopping time for early classification of time series.<n>In finite horizon scenarios, where input lengths are finite, determining the optimal stopping rule becomes computationally intensive due to the need for backward induction.<n>We introduce FIRMBOUND, an SPRT-based framework that efficiently estimates the solution to backward induction from training data.
  arXiv  Detail & Related papers  (2025-01-29T23:54:46Z)
• Distributed Stochastic Gradient Descent with Staleness: A Stochastic Delay Differential Equation Based Framework [56.82432591933544]
  Distributed gradient descent (SGD) has attracted considerable recent attention due to its potential for scaling computational resources, reducing training time, and helping protect user privacy in machine learning.<n>This paper presents the run time and staleness of distributed SGD based on delay differential equations (SDDEs) and the approximation of gradient arrivals.<n>It is interestingly shown that increasing the number of activated workers does not necessarily accelerate distributed SGD due to staleness.
  arXiv  Detail & Related papers  (2024-06-17T02:56:55Z)
• Importance of Smoothness Induced by Optimizers in FL4ASR: Towards Understanding Federated Learning for End-to-End ASR [12.108696564200052]
  We start by training End-to-End Automatic Speech Recognition (ASR) models using Federated Learning (FL) We examine the fundamental considerations that can be pivotal in minimizing the performance gap in terms of word error rate between models trained using FL versus their centralized counterpart.
  arXiv  Detail & Related papers  (2023-09-22T17:23:01Z)
• DFedADMM: Dual Constraints Controlled Model Inconsistency for Decentralized Federated Learning [52.83811558753284]
  Decentralized learning (DFL) discards the central server and establishes a decentralized communication network. Existing DFL methods still suffer from two major challenges: local inconsistency and local overfitting.
  arXiv  Detail & Related papers  (2023-08-16T11:22:36Z)
• Blockchain-enabled Server-less Federated Learning [5.065631761462706]
  We focus on an asynchronous server-less Federated Learning solution empowered by (BC) technology. In contrast to mostly adopted FL approaches, we advocate an asynchronous method whereby model aggregation is done as clients submit their local updates.
  arXiv  Detail & Related papers  (2021-12-15T07:41:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.