TY - JOUR
T1 - SDN Controllers
T2 - A Comprehensive Analysis and Performance Evaluation Study
AU - Zhu, Liehuang
AU - Karim, Md M.
AU - Sharif, Kashif
AU - Xu, Chang
AU - Li, Fan
AU - Du, Xiaojiang
AU - Guizani, Mohsen
N1 - Publisher Copyright:
© 2020 ACM.
PY - 2021/2
Y1 - 2021/2
N2 - Software-defined networks offer flexible and intelligent network operations by splitting a traditional network into a centralized control plane and a programmable data plane. The controller in the control plane is the fundamental element used to manage all operations of the data plane. Hence, the performance and capabilities of the controller itself are essential in achieving optimal performance. Furthermore, the tools used to benchmark their performance must be accurate and useful in measuring different evaluation parameters. There are dozens of controller proposals for general and specialized networks in the literature. However, there is a very limited comprehensive quantitative analysis for them. In this article, we present a comprehensive qualitative comparison of different SDN controllers, along with a quantitative analysis of their performance in different network scenarios. We categorize and classify 34 controllers and present a qualitative comparison. We also present a comparative analysis of controllers for specialized networks such as the Internet of Things, blockchain networks, vehicular networks, and wireless sensor networks. We also discuss in-depth capabilities of benchmarking tools and provide a comparative analysis of their capabilities. This work uses three benchmarking tools to compare 9 controllers and presents a detailed analysis of their performance, along with discussion on performance of specialized network controllers.
AB - Software-defined networks offer flexible and intelligent network operations by splitting a traditional network into a centralized control plane and a programmable data plane. The controller in the control plane is the fundamental element used to manage all operations of the data plane. Hence, the performance and capabilities of the controller itself are essential in achieving optimal performance. Furthermore, the tools used to benchmark their performance must be accurate and useful in measuring different evaluation parameters. There are dozens of controller proposals for general and specialized networks in the literature. However, there is a very limited comprehensive quantitative analysis for them. In this article, we present a comprehensive qualitative comparison of different SDN controllers, along with a quantitative analysis of their performance in different network scenarios. We categorize and classify 34 controllers and present a qualitative comparison. We also present a comparative analysis of controllers for specialized networks such as the Internet of Things, blockchain networks, vehicular networks, and wireless sensor networks. We also discuss in-depth capabilities of benchmarking tools and provide a comparative analysis of their capabilities. This work uses three benchmarking tools to compare 9 controllers and presents a detailed analysis of their performance, along with discussion on performance of specialized network controllers.
KW - Internet of Things
KW - SDN controller
KW - Software-defined networks
KW - benchmarking tools
KW - blockchain
KW - vehicular networks
KW - wireless sensor networks
UR - http://www.scopus.com/inward/record.url?scp=85100791341&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85100791341&partnerID=8YFLogxK
U2 - 10.1145/3421764
DO - 10.1145/3421764
M3 - Review article
AN - SCOPUS:85100791341
SN - 0360-0300
VL - 53
JO - ACM Computing Surveys
JF - ACM Computing Surveys
IS - 6
M1 - 133
ER -