Adaptive Configuration Selection for Multi-Model Inference Pipelines in Edge Computing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sheng, Jinhao, Tang, Zhiqing, Guo, Jianxiong, Wang, Tian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910985767354368
author Sheng, Jinhao
Tang, Zhiqing
Guo, Jianxiong
Wang, Tian
author_facet Sheng, Jinhao
Tang, Zhiqing
Guo, Jianxiong
Wang, Tian
contents The growing demand for real-time processing tasks is driving the need for multi-model inference pipelines on edge devices. However, cost-effectively deploying these pipelines while optimizing Quality of Service (QoS) and costs poses significant challenges. Existing solutions often neglect device resource constraints, focusing mainly on inference accuracy and cost efficiency. To address this, we develop a framework for configuring multi-model inference pipelines. Specifically: 1) We model the decision-making problem by considering the pipeline's QoS, costs, and device resource limitations. 2) We create a feature extraction module using residual networks and a load prediction model based on Long Short-Term Memory (LSTM) to gather comprehensive node and pipeline status information. Then, we implement a Reinforcement Learning (RL) algorithm based on policy gradients for online configuration decisions. 3) Experiments conducted in a real Kubernetes cluster show that our approach significantly improve QoS while reducing costs and shorten decision-making time for complex pipelines compared to baseline algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2506_02814
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adaptive Configuration Selection for Multi-Model Inference Pipelines in Edge Computing
Sheng, Jinhao
Tang, Zhiqing
Guo, Jianxiong
Wang, Tian
Distributed, Parallel, and Cluster Computing
The growing demand for real-time processing tasks is driving the need for multi-model inference pipelines on edge devices. However, cost-effectively deploying these pipelines while optimizing Quality of Service (QoS) and costs poses significant challenges. Existing solutions often neglect device resource constraints, focusing mainly on inference accuracy and cost efficiency. To address this, we develop a framework for configuring multi-model inference pipelines. Specifically: 1) We model the decision-making problem by considering the pipeline's QoS, costs, and device resource limitations. 2) We create a feature extraction module using residual networks and a load prediction model based on Long Short-Term Memory (LSTM) to gather comprehensive node and pipeline status information. Then, we implement a Reinforcement Learning (RL) algorithm based on policy gradients for online configuration decisions. 3) Experiments conducted in a real Kubernetes cluster show that our approach significantly improve QoS while reducing costs and shorten decision-making time for complex pipelines compared to baseline algorithms.
title Adaptive Configuration Selection for Multi-Model Inference Pipelines in Edge Computing
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2506.02814