An incremental flow-based task scheduler for simulation tasks in cloud environments

Miao Zhang; Yong Peng; Quanjun Yin; Qinglong Chen

doi:10.1117/12.2627785

14 March 2022 An incremental flow-based task scheduler for simulation tasks in cloud environments

Miao Zhang, Yong Peng, Quanjun Yin, Qinglong Chen

Proceedings Volume 12165, International Conference on Intelligent Traffic Systems and Smart City (ITSSC 2021); 1216515 (2022) https://doi.org/10.1117/12.2627785
Event: 2021 International Conference on Intelligent Traffic Systems and Smart City, 2021, Zhengzhou, China

Abstract

Deploying distributed simulations on the cloud can obtain many benefits, including lower cost, higher efficiency, easier access and so on. However, due to the virtualization technology adopted by cloud computing, improper assignments of tightly coupled simulation tasks may degrade the overall user experience. At the same time, considering the frequent changes of the status of tasks and hosts, the scheduler should be able to give a good enough solution in time. In this paper, we mainly focus on the efficient scheduling of simulation tasks in the cloud, which has been recognized as an NP-hard combinational optimization problem. Besides mapping such a task-host matching problem as a min-cost max-flow problem, we also design an incremental flow-based task scheduler to deal with the dynamic changes of tasks and hosts. Simulation experiments on Alibaba cluster trace show that our design is adequate to this scenario.

Citation Download Citation

Miao Zhang, Yong Peng, Quanjun Yin, and Qinglong Chen "An incremental flow-based task scheduler for simulation tasks in cloud environments", Proc. SPIE 12165, International Conference on Intelligent Traffic Systems and Smart City (ITSSC 2021), 1216515 (14 March 2022); https://doi.org/10.1117/12.2627785

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available