An SDN-based algorithm for caching, routing, and load balancing in ICN

One of the challenges of Information-centric Networking (ICN) is finding the optimal location for caching content and processing users' requests. In this paper, we address this challenge by leveraging Software-defined Networking (SDN) for efficient ICN management. To achieve this, we formulate...

Full description

Saved in:
Bibliographic Details
Published in:China communications Vol. 21; no. 5; pp. 64 - 76
Main Authors: Tavasoli, MohammadBagher, Saidi, Hossein, Ghiasian, Ali
Format: Journal Article
Language:English
Published: China Institute of Communications 01.05.2024
Department of Electrical and Computer Engineering,Isfahan University of Technology,Isfahan,Iran%Department of Technology and Engineering,Shahrekord University,Shahrekord,Iran
Subjects:
Computer architecture
in-network caching
information-centric network
IP networks
Load management
Load modeling
Network topology
power efficiency
Routing
Servers
software-defined networking
routing
software-defined networking
information-centric network
in-network caching
power efficiency
ISSN:1673-5447
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:One of the challenges of Information-centric Networking (ICN) is finding the optimal location for caching content and processing users' requests. In this paper, we address this challenge by leveraging Software-defined Networking (SDN) for efficient ICN management. To achieve this, we formulate the problem as a mixed-integer nonlinear programming (MINLP) model, incorporating caching, routing, and load balancing decisions. We explore two distinct scenarios to tackle the problem. Firstly, we solve the problem in an offline mode using the GAMS environment, assuming a stable network state to demonstrate the superior performance of the cache-enabled network compared to non-cache networks. Subsequently, we investigate the problem in an online mode where the network state dynamically changes over time. Given the computational complexity associated with MINLP, we propose the software-defined caching, routing, and load balancing (SDCRL) algorithm as an efficient and scalable solution. Our evaluation demonstrates that the SDCRL algorithm significantly reduces computational time while maintaining results that closely resemble those achieved by GAMS.
ISSN:1673-5447
DOI:10.23919/JCC.ja.2023-0165