Paper
Multi-Layer Scheduling for MoE-Based LLM Reasoning
Authors
Yifan Sun, Gholamreza Haffar, Minxian Xu, Rajkumar Buyya, Adel N. Toosi
Abstract
Large Language Models (LLMs) have achieved remarkable success across a wide range of tasks, but serving them efficiently at scale remains a critical challenge due to their substantial computational and latency demands. While most existing inference frameworks rely on simple scheduling strategies such as First-Come-First-Serve (FCFS) at the engine level and Round-Robin (RR) at the scheduler or coordinator level, they often fail to fully utilize system resources and may suffer from issues such as head-of-line blocking and load imbalance. Recent advances in Mixture-of-Experts (MoE) models have also introduced new challenges in scheduling arising from expert parallelism and routing complexity. This research proposes a multi-layer scheduling framework tailored for MoE-based LLM serving. It targets scheduling at three levels: request-level, enginelevel, and expert-level. At the request level, we explore algorithms such as Shortest-Job-First (SJF) and priority-aware aging to improve throughput and reduce latency. At the engine level, we design load-aware dispatching strategies that account for the current prefix token load, KV cache utilization, and user stickiness to achieve better resource matching. At the expert level, we focus on alleviating expert hotspots and strategically placing inter-layer expert dependencies to balance load and improve routing efficiency. Extensive experimental results from more than 100 experiments conducted under diverse workload distributions show that our approach consistently outperforms the state-of-theart inference framework vLLM, achieving up to 17.8% reduction in Time To First Token (TTFT) latency and 13.3% reduction in Time-Per-Output-Token (TPOT) latency.
Metadata
Related papers
Vibe Coding XR: Accelerating AI + XR Prototyping with XR Blocks and Gemini
Ruofei Du, Benjamin Hersh, David Li, Nels Numan, Xun Qian, Yanhe Chen, Zhongy... • 2026-03-25
Comparing Developer and LLM Biases in Code Evaluation
Aditya Mittal, Ryan Shar, Zichu Wu, Shyam Agarwal, Tongshuang Wu, Chris Donah... • 2026-03-25
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
Biplab Pal, Santanu Bhattacharya • 2026-03-25
Retrieval Improvements Do Not Guarantee Better Answers: A Study of RAG for AI Policy QA
Saahil Mathur, Ryan David Rittner, Vedant Ajit Thakur, Daniel Stuart Schiff, ... • 2026-03-25
MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
Zhuo Li, Yupeng Zhang, Pengyu Cheng, Jiajun Song, Mengyu Zhou, Hao Li, Shujie... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2602.21626v1</id>\n <title>Multi-Layer Scheduling for MoE-Based LLM Reasoning</title>\n <updated>2026-02-25T06:42:08Z</updated>\n <link href='https://arxiv.org/abs/2602.21626v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2602.21626v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>Large Language Models (LLMs) have achieved remarkable success across a wide range of tasks, but serving them efficiently at scale remains a critical challenge due to their substantial computational and latency demands. While most existing inference frameworks rely on simple scheduling strategies such as First-Come-First-Serve (FCFS) at the engine level and Round-Robin (RR) at the scheduler or coordinator level, they often fail to fully utilize system resources and may suffer from issues such as head-of-line blocking and load imbalance. Recent advances in Mixture-of-Experts (MoE) models have also introduced new challenges in scheduling arising from expert parallelism and routing complexity. This research proposes a multi-layer scheduling framework tailored for MoE-based LLM serving. It targets scheduling at three levels: request-level, enginelevel, and expert-level. At the request level, we explore algorithms such as Shortest-Job-First (SJF) and priority-aware aging to improve throughput and reduce latency. At the engine level, we design load-aware dispatching strategies that account for the current prefix token load, KV cache utilization, and user stickiness to achieve better resource matching. At the expert level, we focus on alleviating expert hotspots and strategically placing inter-layer expert dependencies to balance load and improve routing efficiency. Extensive experimental results from more than 100 experiments conducted under diverse workload distributions show that our approach consistently outperforms the state-of-theart inference framework vLLM, achieving up to 17.8% reduction in Time To First Token (TTFT) latency and 13.3% reduction in Time-Per-Output-Token (TPOT) latency.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.DC'/>\n <published>2026-02-25T06:42:08Z</published>\n <arxiv:comment>12 pages, 10 figures</arxiv:comment>\n <arxiv:primary_category term='cs.DC'/>\n <author>\n <name>Yifan Sun</name>\n </author>\n <author>\n <name>Gholamreza Haffar</name>\n </author>\n <author>\n <name>Minxian Xu</name>\n </author>\n <author>\n <name>Rajkumar Buyya</name>\n </author>\n <author>\n <name>Adel N. Toosi</name>\n </author>\n </entry>"
}