Saved in:
Bibliographic Details
Main Authors: Zhou, Hanlin, Chan, Huah Yong, Zhang, Shun Yao, Lin, Meie, Ni, Jingfei
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.13579
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914265736151040
author Zhou, Hanlin
Chan, Huah Yong
Zhang, Shun Yao
Lin, Meie
Ni, Jingfei
author_facet Zhou, Hanlin
Chan, Huah Yong
Zhang, Shun Yao
Lin, Meie
Ni, Jingfei
contents With the rise of cloud computing and lightweight containers, Docker has emerged as a leading technology for rapid service deployment, with Kubernetes responsible for pod orchestration. However, for compute-intensive workloads-particularly web services executing containerized machine-learning training-the default Kubernetes scheduler does not always achieve optimal placement. To address this, we propose two custom, reinforcement-learning-based schedulers, SDQN and SDQN-n, both built on the Deep Q-Network (DQN) framework. In compute-intensive scenarios, these models outperform the default Kubernetes scheduler as well as Transformer-and LSTM-based alternatives, reducing average CPU utilization per cluster node by 10%, and by over 20% when using SDQN-n. Moreover, our results show that SDQN-n approach of consolidating pods onto fewer nodes further amplifies resource savings and helps advance greener, more energy-efficient data centers.Therefore, pod scheduling must employ different strategies tailored to each scenario in order to achieve better performance.Since the reinforcement-learning components of the SDQN and SDQN-n architectures proposed in this paper can be easily tuned by adjusting their parameters, they can accommodate the requirements of various future scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2601_13579
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Kubernetes custom scheduler based on reinforcement learning for compute-intensive pods
Zhou, Hanlin
Chan, Huah Yong
Zhang, Shun Yao
Lin, Meie
Ni, Jingfei
Distributed, Parallel, and Cluster Computing
With the rise of cloud computing and lightweight containers, Docker has emerged as a leading technology for rapid service deployment, with Kubernetes responsible for pod orchestration. However, for compute-intensive workloads-particularly web services executing containerized machine-learning training-the default Kubernetes scheduler does not always achieve optimal placement. To address this, we propose two custom, reinforcement-learning-based schedulers, SDQN and SDQN-n, both built on the Deep Q-Network (DQN) framework. In compute-intensive scenarios, these models outperform the default Kubernetes scheduler as well as Transformer-and LSTM-based alternatives, reducing average CPU utilization per cluster node by 10%, and by over 20% when using SDQN-n. Moreover, our results show that SDQN-n approach of consolidating pods onto fewer nodes further amplifies resource savings and helps advance greener, more energy-efficient data centers.Therefore, pod scheduling must employ different strategies tailored to each scenario in order to achieve better performance.Since the reinforcement-learning components of the SDQN and SDQN-n architectures proposed in this paper can be easily tuned by adjusting their parameters, they can accommodate the requirements of various future scenarios.
title A Kubernetes custom scheduler based on reinforcement learning for compute-intensive pods
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2601.13579