Echo: Efficient Co-Scheduling of Hybrid Online-Offline Tasks for Large Language Model Serving

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Zhibin, Li, Shipeng, Li, Xue, Zhou, Yuhang, Zhang, Zhonghui, Wang, Zibo, Gu, Rong, Tian, Chen, Yang, Kun, Zhong, Sheng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916674659155968
author Wang, Zhibin
Li, Shipeng
Li, Xue
Zhou, Yuhang
Zhang, Zhonghui
Wang, Zibo
Gu, Rong
Tian, Chen
Yang, Kun
Zhong, Sheng
author_facet Wang, Zhibin
Li, Shipeng
Li, Xue
Zhou, Yuhang
Zhang, Zhonghui
Wang, Zibo
Gu, Rong
Tian, Chen
Yang, Kun
Zhong, Sheng
contents Large language models have been widely deployed in various applications, encompassing both interactive online tasks and batched offline tasks. Given the burstiness and latency sensitivity of online tasks, over-provisioning resources is common practice. This allows for the integration of latency-insensitive offline tasks during periods of low online load, enhancing resource utilization. However, strategically serving online and offline tasks through a preemption mechanism fails to fully leverage the flexibility of offline tasks and suffers from KV cache recomputation and irregular workloads. In this paper, we introduce Echo, a collaborative online-offline task serving system, including a scheduler, a KV cache manager, and estimation toolkits. The scheduler and KV cache manager work tightly to maximize the throughput of offline tasks, while the estimator further predicts execution time to ensure online task SLOs. The scheduler leverages the batch information of last iteration to reduce the search space for finding the optimal schedule. The KV cache manager sets the priority of the KV cache based on the type of tasks and the opportunity of prefix sharing to reduce the recomputation. Finally, the estimation toolkits predict the execution time, future memory consumption, and the throughput of offline tasks to guide the scheduler, KV cache manager, and the system deployer. Evaluation based on real-world workloads demonstrates that Echo can increase offline task throughput by up to $3.3\times$, while satisfying online task SLOs.
format Preprint
id arxiv_https___arxiv_org_abs_2504_03651
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Echo: Efficient Co-Scheduling of Hybrid Online-Offline Tasks for Large Language Model Serving
Wang, Zhibin
Li, Shipeng
Li, Xue
Zhou, Yuhang
Zhang, Zhonghui
Wang, Zibo
Gu, Rong
Tian, Chen
Yang, Kun
Zhong, Sheng
Distributed, Parallel, and Cluster Computing
Artificial Intelligence
Machine Learning
Large language models have been widely deployed in various applications, encompassing both interactive online tasks and batched offline tasks. Given the burstiness and latency sensitivity of online tasks, over-provisioning resources is common practice. This allows for the integration of latency-insensitive offline tasks during periods of low online load, enhancing resource utilization. However, strategically serving online and offline tasks through a preemption mechanism fails to fully leverage the flexibility of offline tasks and suffers from KV cache recomputation and irregular workloads. In this paper, we introduce Echo, a collaborative online-offline task serving system, including a scheduler, a KV cache manager, and estimation toolkits. The scheduler and KV cache manager work tightly to maximize the throughput of offline tasks, while the estimator further predicts execution time to ensure online task SLOs. The scheduler leverages the batch information of last iteration to reduce the search space for finding the optimal schedule. The KV cache manager sets the priority of the KV cache based on the type of tasks and the opportunity of prefix sharing to reduce the recomputation. Finally, the estimation toolkits predict the execution time, future memory consumption, and the throughput of offline tasks to guide the scheduler, KV cache manager, and the system deployer. Evaluation based on real-world workloads demonstrates that Echo can increase offline task throughput by up to $3.3\times$, while satisfying online task SLOs.
title Echo: Efficient Co-Scheduling of Hybrid Online-Offline Tasks for Large Language Model Serving
topic Distributed, Parallel, and Cluster Computing
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2504.03651