# Stage-Aware Communication Scheduling for Disaggregated LLM Serving

arXiv: https://arxiv.org/abs/2603.17456 · HTML: https://mlsanity.com/paper/2603.17456/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Serving systems and request scheduling, one of 12 directions in Efficient LLM inference and serving.
- Builds on 31 of its 98 references that are in the map.
- First posted on arXiv 18 Mar 2026.

## Placed in

- Efficient LLM inference and serving / [Serving systems and request scheduling](https://mlsanity.com/inference/serving-systems-and-request-scheduling/)

## Builds on (most cited first, 30 of 31)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM](https://mlsanity.com/paper/2104.04473/) (2021)
- [Splitwise: Efficient generative LLM inference using phase splitting](https://mlsanity.com/paper/2311.18677/) (2023)
- [Ring Attention with Blockwise Transformers for Near-Infinite Context](https://mlsanity.com/paper/2310.01889/) (2023)
- [DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale](https://mlsanity.com/paper/2207.00032/) (2022)
- [Multi-Agent Collaboration: Harnessing the Power of Intelligent LLM Agents](https://mlsanity.com/paper/2306.03314/) (2023)
- [CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion](https://mlsanity.com/paper/2405.16444/) (2024)
- [CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Serving](https://mlsanity.com/paper/2310.07240/) (2023)
- [Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving](https://mlsanity.com/paper/2407.00079/) (2024)
- [Inference without Interference: Disaggregate LLM Inference for Mixed Downstream Workloads](https://mlsanity.com/paper/2401.11181/) (2024)
- [Vidur: A Large-Scale Simulation Framework For LLM Inference](https://mlsanity.com/paper/2405.05465/) (2024)
- [LoongServe: Efficiently Serving Long-Context Large Language Models with Elastic Sequence Parallelism](https://mlsanity.com/paper/2404.09526/) (2024)
- [Accelerating Distributed MoE Training and Inference with Lina](https://mlsanity.com/paper/2210.17223/) (2022)
- [Can ChatGPT replace StackOverflow? A Study on Robustness and Reliability of Large Language Model Code Generation](https://mlsanity.com/paper/2308.10335/) (2023)
- [Qwen3-Coder-Next Technical Report](https://mlsanity.com/paper/2603.00729/) (2026)
- [MemServe: Context Caching for Disaggregated LLM Serving with Elastic Memory Pool](https://mlsanity.com/paper/2406.17565/) (2024)
- [TACCL: Guiding Collective Algorithm Synthesis using Communication Sketches](https://mlsanity.com/paper/2111.04867/) (2021)
- [CASSINI: Network-Aware Job Scheduling in Machine Learning Clusters](https://mlsanity.com/paper/2308.00852/) (2023)
- [Context Parallelism for Scalable Million-Token Inference](https://mlsanity.com/paper/2411.01783/) (2024)
- [Triton-distributed: Programming Overlapping Kernels on Distributed AI Systems with the Triton Compiler](https://mlsanity.com/paper/2504.19442/) (2025)
- [TileLink: Generating Efficient Compute-Communication Overlapping Kernels using Tile-Centric Primitives](https://mlsanity.com/paper/2503.20313/) (2025)
- [ThunderServe: High-performance and Cost-efficient LLM Serving in Cloud Environments](https://mlsanity.com/paper/2502.09334/) (2025)
- [SLOs-Serve: Optimized Serving of Multi-SLO LLMs](https://mlsanity.com/paper/2504.08784/) (2025)
- [DualPath: Breaking the Storage Bandwidth Bottleneck in Agentic LLM Inference](https://mlsanity.com/paper/2602.21548/) (2026)
- [TraCT: Disaggregated LLM Serving with CXL Shared Memory KV Cache at Rack-Scale](https://mlsanity.com/paper/2512.18194/) (2025)
- [FlowKV: A Disaggregated Inference Framework with Low-Latency KV Cache Transfer and Load-Aware Scheduling](https://mlsanity.com/paper/2504.03775/) (2025)
- [MixNet: A Runtime Reconfigurable Optical-Electrical Fabric for Distributed Mixture-of-Experts Training](https://mlsanity.com/paper/2501.03905/) (2025)
- [Toward Low-Latency End-to-End Voice Agents for Telecommunications Using Streaming ASR, Quantized LLMs, and Real-Time TTS](https://mlsanity.com/paper/2508.04721/) (2025)
- [Enhancing reliability in AI inference services: An empirical study on real production incidents](https://mlsanity.com/paper/2511.07424/) (2025)

## Built on by (most cited first, 1 of 1)

- [MOSAIC: Masked Outsourcing of Secure AI Computations](https://mlsanity.com/paper/2607.29221/) (2026)

## Abstract (arXiv)

Meeting stringent Time-To-First-Token (TTFT) requirements is crucial for LLM applications. To improve efficiency, modern LLM serving systems adopt disaggregated architectures with diverse parallelisms, introducing complex multi-stage workflows involving reusable KV-block retrieval, collective communication, and P2D transfer. Flows from dependent stages overlap within and across requests on shared bottleneck links, making TTFT highly susceptible to network contention and necessitating stage-aware scheduling. Unfortunately, most existing works schedule flows in a stage-agnostic manner, leading to uncoordinated contention that constitutes a primary cause of SLO violations. In this paper, we present Nuska, a holistic multi-stage flow scheduling mechanism designed to maximize TTFT SLO attainment. At its core, Nuska approximates the Least-Laxity-First (LLF) scheduling policy without requiring precise knowledge of a request's remaining slack. It achieves this through a Defer-and-Promote principle implemented through a Reverse Multi-Level Queue (RMLQ) structure. By dynamically promoting task precedence as effective laxity diminishes, Nuska prioritizes flows with less laxity while preventing requests with loose SLOs from prematurely consuming network bandwidth. We implement Nuska as a pluggable module integrated into vLLM, and evaluate it on an 8-server, 32-GPU testbed as well as through large-scale simulations. Our results demonstrate that Nuska effectively outperforms state-of-the-art baselines, improving the TTFT SLO attainment by 1.2x-2.4x.