Seminar: "End-to-End Performance-based Autonomous VNF Placement with adopted Reinforcement Learning"

16 June 2020, 10.30 AM - 16 June 2020, 11.30 AM

Mr Monchai Bunyakitanon & Dr Xenofon Vasilakos

Speakers: Mr Monchai Bunyakitanon & Dr Xenofon Vasilakos 
Title: "End-to-End Performance-based Autonomous VNF Placement with adopted Reinforcement Learning"
 
Abstract:
 
The autonomous placement of Virtual Network Functions (VNFs) is a key aspect of Zero-touch network and Service Management (ZSM) in Fifth Generation (5G) networking. Therefore, current orchestration frameworks need to be enhanced, accordingly. To address this need, this work presents an Adapted REinforcement Learning VNF Performance Prediction module for Autonomous VNF Placement, namely AREL3P. Our solution design bears a dual novelty. First, it leverages end-to-end service-level performance predictions for placing VNFs. Second, whereas the majority of other Machine Learning efforts in the literature use Supervised Learning (SL) techniques, AREL3P is based on a particular form of Reinforcement Learning adapted to predictions. This makes placement decisions more resilient to dynamic conditions, as well as portable to other network nodes, and able to generalize in heterogeneous network environments. Backed by a meticulous performance evaluation over a real 5G end-to-end testbed, we verify the above properties after integrating AREL3P to Open Source Management and Orchestration (OSM MANO) decisions. Among other highlights, we show increased VNF performance predictions accuracy by 40 - 45%, and an overall improved VNF placement efficiency against other SL benchmarks reflected by near-optimal decision scores in 23 out of a total of 27 investigated scenarios.
 
Monchai Bio: 
 
Mr. Bunyakitanon earned his BSEE degree from Escuela Naval Militar, Pontevedra, Spain, and received his MSEE degree with an emphasis in Telecommunication System from Blekinge Institute of Technology, Karlskrona, Sweden. Monchai is currently a Ph.D. student in the High Performance Network research group at the University of Bristol.
 
Xenofon Bio: 
 
Dr. Xenofon Vasilakos is a Research Fellow with the University of Bristol and a member of the Smart Internet Lab and the High Performance Networks research group. His current and past research involves 5G and beyond architectures, concepts and technologies including SDN/NFV, Pervasive Networking, and Network Slicing, with particular interest on Multi-access Edge Computing (MEC) solutions combined to intelligent/cognitive approaches inspired by Machine Learning (ML) models towards zero-touch self-managed networks. Since he joined the Smart Internet Lab, he is working for the EU H2020 5GINFire research project. In the past, he has worked for various EU funded H2020 and FP7 research projects such as 5GPPP-SliceNet and the FIA award-winning FP7-PURSUIT.

Past experience: His recent past research work involved 5G slicing technologies with a particular focus on MEC solutions and cognition approaches inspired by ML models towards self-managed 5G mobile networks. During his Ph.D. studies, he conducted research on clean-slate Information-Centric Networking (ICN) architectures, protocols and distributed solutions for the Future Internet, with an emphasis on proactive caching-based seamless mobility support, and rendezvous-based name resolution. He has co-supervised student theses and worked as a teaching assistant or teacher on various M.Sc. and B.Sc. courses such as Distributed Systems, Operating Systems, Android programming and Java programming. He is a regular reviewer for top-tier conferences and journals.

Educational background: Dr. Vasilakos holds a Ph.D. in Informatics from the Athens University of Economics and Business (AUEB), an M.Sc. degree in Parallel and Distributed Computer Systems from the Vrije Universiteit of Amsterdam and a B.Sc. in Informatics from AUEB. Before joining the Smart Internet Lab, he served as a postdoctoral fellow at Eurecom graduate school & research center, awarded with the French government LABEX (LABoratoires d’EXcellence) Postdoc Grant. While at Eurecom, he was also a member of the Mosaic 5G team for building open source community-led agile 5G platforms and projects.

 

 
 
 

Contact information

For more information about this event, please contact Harriet Parks, Smart Internet Lab Project Coordinator - 0117 3315356 or Harriet.Parks@bristol.ac.uk

Mr Monchai Bunyakitanon

Dr Xenofon Vasilakos

Edit this page