A Software Campus project
RealVNF is a 2-year project (11/2018 to 01/2021) that is funded by the German Ministry for Education and Research (BMBF) and part of the Software Campus. The project is a collaboration between Paderborn University, Germany, and Huawei Munich.
The goal of the project is to design and develop concepts and algorithms to improve coordination of chained virtual network functions (VNFs) under realistic conditions, take cloud computing and network softwarization to the next level. A strong focus is on novel approaches leveraging machine learning and reinforcement learning.
Official RealVNF project description: Software Campus RealVNF website
Related blog post: Lessons learned from leading my first project
- Stefan Schneider, Paderborn University
- Artur Hecker, Huawei Munich Research Center
- Ramin Khalili, Huawei Munich Research Center
Student researchers and developers:
- Adnan Manzoor, Paderborn University
- Haydar Qarawlus, Paderborn University
- Rafael Schellenberg, Paderborn University
- Sven Uthe, Paderborn University
Distributed Online Service Coordination Using Deep Reinforcement Learning S. Schneider, H. Qarawlus, and H. Karl IEEE International Conference on Distributed Computing Systems (ICDCS), 2021 [preprint] [code]
Self-Driving Network and Service Coordination Using Deep Reinforcement Learning S. Schneider, A. Manzoor, H. Qarawlus, R. Schellenberg, H. Karl, R. Khalili, and A. Hecker IEEE International Conference on Network and Service Management (CNSM), 2020 [preprint] [code]
Best Student Paper at CNSM 2020
Machine Learning for Dynamic Resource Allocation in Network Function Virtualization S. Schneider, N. P. Satheeschandran, M. Peuster, and H. Karl IEEE Conference on Network Softwarization (NetSoft), 2020 [preprint] [code]
- deep-rl-network-service-coordination: A deep reinforcement learning framework for self-driving network & service coordination.
- coord-sim: A Python lightweight flow-level simulator for evaluating network & service coordination algorithms.
- hierarchical-coordination: Hierarchical network & service coordination using a divide-and-conquer strategy
- distributed-coordination: Fully distributed algorithms for highly scalable network & service coordination.
- ml-for-resource-allocation: Machine learning framework for dynamic resource allocation in NFV.
All repositories: https://github.com/RealVNF/
Self-Learning Network and Service Coordination (Demo):
Self-Driving Network and Service Coordination Using Deep Reinforcement Learning (Conference Talk at IEEE CNSM 2020):
Fully Distributed Service Coordination (Conference Talk at IEEE CNSM 2020):