| Presentation | 2022-04-15 Service Chaining Based on Capacitated Shortest Path Tour Problem Takanori Hara, Masahiro Sasabe, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | The service chaining problem is one of the resource allocation problems in network functions virtualization (NFV) networks and aims at finding a service path from an origin node to a destination node while executing the virtual network functions at intermediate nodes in required order under the resource constraints. In our previous work, we modeled the service chaining problem as the capacitated shortest path tour problem (CSPTP) and formulated the CSPTP-based service chaining as an integer linear program. We also proposed a simple greedy algorithm and more sophisticated Lagrangian heuristics to efficiently solve the CSPTP-based service chaining. In this paper, we aim at realizing CSPTP-based service chaining that can learn demand trend and adapt to topology changes due to link failures by using both deep reinforcement learning and graph neural network. Through numerical results, we demonstrate the fundamental characteristics of the proposed scheme and adaptability to topology change with link failures. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Network functions virtualization / service chaining / capacitated shortest path tour problem / deep reinforcement learning / graph neural network |
| Paper # | NS2022-2 |
| Date of Issue | 2022-04-08 (NS) |
| Conference Information | |
| Committee | NS |
|---|---|
| Conference Date | 2022/4/15(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | kikai shinkou kaikan + online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Communication traffic theory, Traffic and quality evaluation, Network performance evaluation, QoS/QoE, Reliability and robustness, Traffic and quality management, AI and machine learning, Network and system operation management, High capacity, low latency, many connections, General |
| Chair | Akihiro Nakao(Univ. of Tokyo) |
| Vice Chair | Tetsuya Oishi(NTT) |
| Secretary | Tetsuya Oishi(NTT) |
| Assistant | Kotaro Mihara(NTT) |
| Paper Information | |
| Registration To | Technical Committee on Network Systems |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Service Chaining Based on Capacitated Shortest Path Tour Problem |
| Sub Title (in English) | Solution Based on Deep Reinforcement Learning and Graph Neural Network |
| Keyword(1) | Network functions virtualization |
| Keyword(2) | service chaining |
| Keyword(3) | capacitated shortest path tour problem |
| Keyword(4) | deep reinforcement learning |
| Keyword(5) | graph neural network |
| 1st Author's Name | Takanori Hara |
| 1st Author's Affiliation | Nara Institute of Science and Technology(NAIST) |
| 2nd Author's Name | Masahiro Sasabe |
| 2nd Author's Affiliation | Nara Institute of Science and Technology(NAIST) |
| Date | 2022-04-15 |
| Paper # | NS2022-2 |
| Volume (vol) | vol.122 |
| Number (no) | NS-5 |
| Page | pp.pp.7-12(NS), |
| #Pages | 6 |
| Date of Issue | 2022-04-08 (NS) |