学术报告
我的位置在: 首页 > 学术报告 > 正文
Game theory based load balancing in distributed systems with applications
浏览次数:日期:2020-12-09编辑:信科院 科研办

报告人:  Anthony T. Chronopoulos,教授,美国德克萨斯大学圣安东尼奥分校。

报告时间:202012月9日 (星期三) 下午4:00 - 6:00

报告地点:Zoom在线会议

https://us02web.zoom.us/j/2810019605?pwd=S09LNnl5dHdXajZBbEJJOVd4TVlmUT09

Meeting ID: 281 001 9605

Passcode: HNU2020

欢迎广大师生参加!烦请大家提前安装Zoom会议软件,软件下载地址:https://zoom.us/download


报告摘要:The Distributed Systems (DS) resources allocation  problem can be modeled using Game Theory (GT), where users (represented by agents) compete for resources. Individual users’ optimization schemes had been proposed in the past to model this problem  (by H. Kameda et al).   However, our group were the first to propose a GT model (with a finite number of players) for this problem inspired by   S. Nash’s GT in economics . We proposed original algorithms to compute the solution (proved to a Nash equilibrium), which are distributed and scalable. The new algorithms improve the system's utilization  and guarantee Quality of Service guarantees (QoS) (‘fair allocation’) in terms of delivered performance to the users.

For DS in which all the jobs belong to a single user (single-class), we used a cooperative game to model the load balancing problem which takes the average system information into account (static load balancing). The solution is based on the Nash Bargaining Solution which provides a Pareto optimal solution for the distributed system and is also a fair solution. Applications of the proposed GT algorithms have been extended  and applied to Cloud and Grid Computing.

 

报告人简介:Dr. Anthony Theodore Chronopoulos obtained a Ph.D. in Computer Science from the University of Illinois at Urbana-Champaign in 1987. He is a full  professor at the Department of Computer Science, University of Texas, San Antonio, USA and a visiting professor, Department of Computer Engineering & Informatics, University of Patras, Greece. He is the author of 95 journal and 73 peer-reviewed conference proceedings publications in the areas of Parallel and Distributed Computing, Grid and Cloud Computing, Machine Learning, Numerical and Scientific Computing, Computer Networks and Wireless Communications, Applications to Science and Engineering. He is a Fellow of the Institution of Engineering and Technology (FIET), ACM Senior member, IEEE Senior member.

 

邀请人:李肯立

 

联系人:陈建国