Using NFV and Reinforcement Learning for Anomalies Detection and Mitigation in SDN

Computer networks are subject to several anomalies, which leads to the necessity of techniques to coordinate detection and mitigation to keep the network operational. In this paper we propose the use of reinforcement learning to promote resilience in Software Defined Networking (SDN). In particular,...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE Symposium on Computers and Communications (ISCC) pp. 00432 - 00437
Main Authors Sampaio, Lauren S. R., Faustini, Pedro H. A., Silva, Anderson S., Granville, Lisandro Z., Schaeffer-Filho, Alberto
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2018
Subjects
anomaly detection
Hardware
machine learning
Network function virtualization
network functions virtualization
Prototypes
reinforcement learning
Servers
Software
software defined networking
Switches
Online AccessGet full text
DOI10.1109/ISCC.2018.8538614

Cover

More Information
Summary:Computer networks are subject to several anomalies, which leads to the necessity of techniques to coordinate detection and mitigation to keep the network operational. In this paper we propose the use of reinforcement learning to promote resilience in Software Defined Networking (SDN). In particular, it is proposed collecting network metrics and grouping them into profiles, each one having a set of actions that handles problems using reinforcement learning, Network Functions Virtualization (NFV), and an SDN controller. Policies for dealing with anomalies are defined based on rewards for each action. Results show that the system obtains mostly positive rewards, but a small increment in the topology size leads to more than four times the number of entries in the state-action table.
DOI:10.1109/ISCC.2018.8538614