Semantic Scheduling for LLM Inference

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hua, Wenyue, Ding, Dujian, Gu, Yile, Ren, Yujie, Mei, Kai, Ma, Minghua, Wang, William Yang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915343267528704
author Hua, Wenyue
Ding, Dujian
Gu, Yile
Ren, Yujie
Mei, Kai
Ma, Minghua
Wang, William Yang
author_facet Hua, Wenyue
Ding, Dujian
Gu, Yile
Ren, Yujie
Mei, Kai
Ma, Minghua
Wang, William Yang
contents Conventional operating system scheduling algorithms are largely content-ignorant, making decisions based on factors such as latency or fairness without considering the actual intents or semantics of processes. Consequently, these algorithms often do not prioritize tasks that require urgent attention or carry higher importance, such as in emergency management scenarios. However, recent advances in language models enable semantic analysis of processes, allowing for more intelligent and context-aware scheduling decisions. In this paper, we introduce the concept of semantic scheduling in scheduling of requests from large language models (LLM), where the semantics of the process guide the scheduling priorities. We present a novel scheduling algorithm with optimal time complexity, designed to minimize the overall waiting time in LLM-based prompt scheduling. To illustrate its effectiveness, we present a medical emergency management application, underscoring the potential benefits of semantic scheduling for critical, time-sensitive tasks. The code and data are available at https://github.com/Wenyueh/latency_optimization_with_priority_constraints.
format Preprint
id arxiv_https___arxiv_org_abs_2506_12204
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Semantic Scheduling for LLM Inference
Hua, Wenyue
Ding, Dujian
Gu, Yile
Ren, Yujie
Mei, Kai
Ma, Minghua
Wang, William Yang
Machine Learning
Artificial Intelligence
Operating Systems
Conventional operating system scheduling algorithms are largely content-ignorant, making decisions based on factors such as latency or fairness without considering the actual intents or semantics of processes. Consequently, these algorithms often do not prioritize tasks that require urgent attention or carry higher importance, such as in emergency management scenarios. However, recent advances in language models enable semantic analysis of processes, allowing for more intelligent and context-aware scheduling decisions. In this paper, we introduce the concept of semantic scheduling in scheduling of requests from large language models (LLM), where the semantics of the process guide the scheduling priorities. We present a novel scheduling algorithm with optimal time complexity, designed to minimize the overall waiting time in LLM-based prompt scheduling. To illustrate its effectiveness, we present a medical emergency management application, underscoring the potential benefits of semantic scheduling for critical, time-sensitive tasks. The code and data are available at https://github.com/Wenyueh/latency_optimization_with_priority_constraints.
title Semantic Scheduling for LLM Inference
topic Machine Learning
Artificial Intelligence
Operating Systems
url https://arxiv.org/abs/2506.12204