About the journal

• Home page
• Contact information
• Aims and scope
• Indexing information
• Editorial policies
• ComSIS consortium
• Journal boards
• Managing board

For authors

• Information for contributors
• Paper submission
• Article submission through OJS
• Copyright transfer form
• Download section

For readers

• Forthcoming articles
• Current issue
• Archive

For reviewers

• View and review submissions

News

• Journal's Facebook page
• Calls for special issues
• New issue notification

A Load Balancing Scheme for Gaming Server applying Reinforcement Learning in IoT

Hye-Young Kim¹ and Jin-Sul Kim²

1. School of Games, Major of Game Software
   Hongik University, Korea
   {hykim@hongik.ac.kr
2. School of Electronics and Computer Engineering
   Chonnam National University, Gwangju, South Kore
   jsworld@jnu.ac.kr

Abstract

A lot of data generated on the game server causes overtime in IoT environment. Recently, both researchers and developers have developed great interests in load balancing schemes in gaming servers. The existing literature have proposed algorithms that distribute loads in servers by mostly concentrating on load balancing and cooperative offloading in Internt of Things (IoT) environment. The dynamic load balancing algorithms have applied a technique of calculating the workload of the network and dynamically allocating the workload according to the network situation, taking into account the capacity of the servers. However, the various previous researches proposed are difficult to reflect the real world by imposing a lot of restrictions and assumptions on the IoT environment, and it is not enough to meet the wide range of service requirements for the IoT environment. Therefore, we proposed an agent that applies a deep reinforced learning method to distribute loads for gaming servers. The agent has accomplished this by measuring network loads and analyzing a large amount of user data. We specifically have chosen deep reinforcement learning because no labels would need to be obtained in advance and it enabled our agent to immediately make the right decisions to load balancing in IoT environment. We have showed several siginicicant functions of our proposed scheme and derived through mathematical analysis. Also, we have compared performances of our proposed scheme and a previus research, ProGreGA, widely used scheme through simulation.

Key words

deep reinforcement learning, load balancing, gaming server, reward, achievable rate, loss rate, policy

Digital Object Identifier (DOI)

https://doi.org/10.2298/CSIS190917026K

Publication information

Volume 17, Issue 3 (October 2020)
Year of Publication: 2020
ISSN: 2406-1018 (Online)
Publisher: ComSIS Consortium

Full text

Available in PDF
Portable Document Format

How to cite

Kim, H., Kim, J.: A Load Balancing Scheme for Gaming Server applying Reinforcement Learning in IoT. Computer Science and Information Systems, Vol. 17, No. 3, 891–906. (2020), https://doi.org/10.2298/CSIS190917026K