Book chapter
New Distributed Constraint Reasoning Algorithms for Load Balancing in Edge Computing
PRIMA 2019: Principles and Practice of Multi-Agent Systems, pp.69-86
Lecture Notes in Computer Science, 11873, Springer International Publishing
10/21/2019
DOI: 10.1007/978-3-030-33792-6_5
Abstract
Edge computing is a paradigm for improving the performance of cloud computing systems by performing data processing at the edge of the network, closer to the users and sources of data. As data processing is traditionally done in large data centers, typically located far from their users, the edge computing paradigm will reduce the communication bottleneck between the user and the location of data processing, thereby improving overall performance. This becomes more important as the number of Internet-of-Things (IoT) devices and other mobile or embedded devices continues to increase. In this paper, we investigate the use of distributed constraint reasoning (DCR) techniques to model and solve the distributed load balancing problem in edge computing problems. Specifically, we (i) provide a mapping of the distributed load balancing problem in edge computing to a distributed constraint satisfaction and optimization problem; (ii) propose two DCR algorithms to solve such problems; and (iii) empirically evaluate our algorithms against a state-of-the-art DCR algorithm on random and scale-free networks.
Details
- Title: Subtitle
- New Distributed Constraint Reasoning Algorithms for Load Balancing in Edge Computing
- Creators
- Khoi D Hoang - Washington University in St. LouisChristabel Wayllace - Washington University in St. LouisWilliam Yeoh - Washington University in St. LouisJacob Beal - University of Iowa, Electrical and Computer EngineeringSoura Dasgupta - University of Iowa, Electrical and Computer EngineeringYuanqiu Mo - Department of Electrical and Computer Engineering, University of Iowa, Iowa City, USAAaron Paulos - Raytheon BBN Technologies, Cambridge, USAJon Schewe - Raytheon BBN Technologies, Cambridge, USA
- Resource Type
- Book chapter
- Publication Details
- PRIMA 2019: Principles and Practice of Multi-Agent Systems, pp.69-86
- Publisher
- Springer International Publishing; Cham
- Series
- Lecture Notes in Computer Science; 11873
- DOI
- 10.1007/978-3-030-33792-6_5
- eISSN
- 1611-3349
- ISSN
- 0302-9743
- Language
- English
- Date published
- 10/21/2019
- Academic Unit
- Electrical and Computer Engineering
- Record Identifier
- 9984197161602771
Metrics
22 Record Views