Irina Bolodurina1 and Denis Parfenov2
1Department of Applied Mathematics Orenburg State University Orenburg, Russia
2
Faculty of Distance Learning Technologies Orenburg State University Orenburg, Russia
ABSTRACT
This paper represents the results of the research, which have allowed us to develop a hybrid
approach to the processing, classification, and control of traffic routes. The approach enables to
identify traffic flows in the virtual data center in real-time systems. Our solution is based on the
methods of data mining and machine learning, which enable to classify traffic more accurately
according to more criteria and parameters. As a practical result, the paper represents the
algorithmic solution of the classification of the traffic flows of cloud applications and services
embodied in a module for the controller of the software-defined network. This solution enables to
increase the efficiency of handling user requests to cloud applications and reduce the response
time, which has a positive effect on the quality of service in the network of the virtual data center.
KEYWORDS
Cloud applications; software-defined network; traffic flows; virtual data center; data mining;
machine learning
No comments:
Post a Comment