Past-Future Scheduler for LLM Serving under SLA Guarantees

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gong, Ruihao, Bai, Shihao, Wu, Siyu, Fan, Yunqian, Wang, Zaijun, Li, Xiuhong, Yang, Hailong, Liu, Xianglong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915388671918080
author Gong, Ruihao
Bai, Shihao
Wu, Siyu
Fan, Yunqian
Wang, Zaijun
Li, Xiuhong
Yang, Hailong
Liu, Xianglong
author_facet Gong, Ruihao
Bai, Shihao
Wu, Siyu
Fan, Yunqian
Wang, Zaijun
Li, Xiuhong
Yang, Hailong
Liu, Xianglong
contents The exploration and application of Large Language Models (LLMs) is thriving. To reduce deployment costs, continuous batching has become an essential feature in current service frameworks. The effectiveness of continuous batching relies on an accurate estimate of the memory requirements of requests. However, due to the diversity in request output lengths, existing frameworks tend to adopt aggressive or conservative schedulers, which often result in significant overestimation or underestimation of memory consumption. Consequently, they suffer from harmful request evictions or prolonged queuing times, failing to achieve satisfactory throughput under strict Service Level Agreement (SLA) guarantees (a.k.a. goodput), across various LLM application scenarios with differing input-output length distributions. To address this issue, we propose a novel Past-Future scheduler that precisely estimates the peak memory resources required by the running batch via considering the historical distribution of request output lengths and calculating memory occupancy at each future time point. It adapts to applications with all types of input-output length distributions, balancing the trade-off between request queuing and harmful evictions, thereby consistently achieving better goodput. Furthermore, to validate the effectiveness of the proposed scheduler, we developed a high-performance LLM serving framework, LightLLM, that implements the Past-Future scheduler. Compared to existing aggressive or conservative schedulers, LightLLM demonstrates superior goodput, achieving up to 2-3$\times$ higher goodput than other schedulers under heavy loads. LightLLM is open source to boost the research in such direction (https://github.com/ModelTC/lightllm).
format Preprint
id arxiv_https___arxiv_org_abs_2507_10150
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Past-Future Scheduler for LLM Serving under SLA Guarantees
Gong, Ruihao
Bai, Shihao
Wu, Siyu
Fan, Yunqian
Wang, Zaijun
Li, Xiuhong
Yang, Hailong
Liu, Xianglong
Distributed, Parallel, and Cluster Computing
The exploration and application of Large Language Models (LLMs) is thriving. To reduce deployment costs, continuous batching has become an essential feature in current service frameworks. The effectiveness of continuous batching relies on an accurate estimate of the memory requirements of requests. However, due to the diversity in request output lengths, existing frameworks tend to adopt aggressive or conservative schedulers, which often result in significant overestimation or underestimation of memory consumption. Consequently, they suffer from harmful request evictions or prolonged queuing times, failing to achieve satisfactory throughput under strict Service Level Agreement (SLA) guarantees (a.k.a. goodput), across various LLM application scenarios with differing input-output length distributions. To address this issue, we propose a novel Past-Future scheduler that precisely estimates the peak memory resources required by the running batch via considering the historical distribution of request output lengths and calculating memory occupancy at each future time point. It adapts to applications with all types of input-output length distributions, balancing the trade-off between request queuing and harmful evictions, thereby consistently achieving better goodput. Furthermore, to validate the effectiveness of the proposed scheduler, we developed a high-performance LLM serving framework, LightLLM, that implements the Past-Future scheduler. Compared to existing aggressive or conservative schedulers, LightLLM demonstrates superior goodput, achieving up to 2-3$\times$ higher goodput than other schedulers under heavy loads. LightLLM is open source to boost the research in such direction (https://github.com/ModelTC/lightllm).
title Past-Future Scheduler for LLM Serving under SLA Guarantees
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2507.10150