An Advanced Reinforcement Learning Framework for Online Scheduling of Deferrable Workloads in Cloud Computing

TL;DR

This paper introduces extit{OSDEC}, a deep reinforcement learning-based online scheduling framework for deferrable workloads in cloud computing, optimizing resource utilization and reducing user waiting time.

Contribution

It presents a novel RL-based online scheduling method for deferrable jobs, incorporating auxiliary tasks for improved state representation and performance.

Findings

Achieves high resource utilization in cloud platforms.

Reduces user waiting time compared to baseline methods.

Validated on a public dataset with superior results.

Abstract

Efficient resource utilization and perfect user experience usually conflict with each other in cloud computing platforms. Great efforts have been invested in increasing resource utilization but trying not to affect users' experience for cloud computing platforms. In order to better utilize the remaining pieces of computing resources spread over the whole platform, deferrable jobs are provided with a discounted price to users. For this type of deferrable jobs, users are allowed to submit jobs that will run for a specific uninterrupted duration in a flexible range of time in the future with a great discount. With these deferrable jobs to be scheduled under the remaining capacity after deploying those on-demand jobs, it remains a challenge to achieve high resource utilization and meanwhile shorten the waiting time for users as much as possible in an online manner. In this paper, we propose an online deferrable job scheduling method called \textit{Online Scheduling for DEferrable jobs in Cloud} (\OSDEC{}), where a deep reinforcement learning model is adopted to learn the scheduling policy, and several auxiliary tasks are utilized to provide better state representations and improve the performance of the model. With the integrated reinforcement learning framework, the proposed method can well plan the deployment schedule and achieve a short waiting time for users while maintaining a high resource utilization for the platform. The proposed method is validated on a public dataset and shows superior performance.