Presentation | 2022-03-10 VNF/CNF migration scheduling based on Encoder-Decoder RNN for cloud native platform Takahiro Hirayama, Masahiro Jibiki, Takaya Miyazawa, Ved P. Kafle, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The 5th generation (5G) or beyond 5G (B5G) mobile networks are required to offer various kinds of services, such as enhanced mobile broadband, ultra-reliable, ultra-low latency, and massive machine type communications, over the same infrastructure. Network function virtualization (NFV) technology enables operators to dynamically allocate computing resources to virtual or container network functions (VNFs and/or CNFs) in response to changes of users’ requests. To meet the quality of service (QoS) requirements even when the demands for resources such as CPU and bandwidth dynamically vary, the NFV system is required to find an optimal solution of migrating NFs from saturated servers to others. To quickly adapt to changes of network conditions, NF migration scheduling should be solved in a short time. However, if the system tries to find the optimal solution of the scheduling problem using the existing integer linear programming (ILP), it hard to obtain the solution in a short time because of the computation complexity. Application of machine learning (ML) technologies is an effective approach to deal with this issue. In this paper, we firstly formulate the NF migration scheduling problem as the ILP. And then, our approach is to make an artificial intelligence (AI) consisting of encoder-decoder recurrent neural network models train the solutions obtained by the above ILP. As a result of simulation, we prove that our approach can agiely provide solutions of NF migration scheduling problems with achieving both mitigation of redundant migration of NFs and reduction of computation time for training. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Service Function Chaining (SFC) / Integer Linear Programming (ILP) / Machine Learning (ML) / Recurrent Neural Network (RNN) / Cloud Native Platform |
Paper # | IN2021-35 |
Date of Issue | 2022-03-03 (IN) |
Conference Information | |
Committee | NS / IN |
---|---|
Conference Date | 2022/3/10(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | General |
Chair | Akihiro Nakao(Univ. of Tokyo) / Kenji Ishida(Hiroshima City Univ.) |
Vice Chair | Tetsuya Oishi(NTT) / Kunio Hato(Internet Multifeed) |
Secretary | Tetsuya Oishi(NTT) / Kunio Hato(Chuo Univ.) |
Assistant | Kotaro Mihara(NTT) |
Paper Information | |
Registration To | Technical Committee on Network Systems / Technical Committee on Information Networks |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | VNF/CNF migration scheduling based on Encoder-Decoder RNN for cloud native platform |
Sub Title (in English) | |
Keyword(1) | Service Function Chaining (SFC) |
Keyword(2) | Integer Linear Programming (ILP) |
Keyword(3) | Machine Learning (ML) |
Keyword(4) | Recurrent Neural Network (RNN) |
Keyword(5) | Cloud Native Platform |
1st Author's Name | Takahiro Hirayama |
1st Author's Affiliation | National Institute of Information and Communications Technology(NICT) |
2nd Author's Name | Masahiro Jibiki |
2nd Author's Affiliation | National Institute of Information and Communications Technology(NICT) |
3rd Author's Name | Takaya Miyazawa |
3rd Author's Affiliation | National Institute of Information and Communications Technology(NICT) |
4th Author's Name | Ved P. Kafle |
4th Author's Affiliation | National Institute of Information and Communications Technology(NICT) |
Date | 2022-03-10 |
Paper # | IN2021-35 |
Volume (vol) | vol.121 |
Number (no) | IN-434 |
Page | pp.pp.25-30(IN), |
#Pages | 6 |
Date of Issue | 2022-03-03 (IN) |