[WP4-A] Poster Session 2

[WP4-A5] Adaptive Failure Prediction Using Long Short-term Memory in Optical Network

〇Chunyu Zhang¹, Minghui Wang², Min Zhang¹, Danshi Wang¹, Chuang Song¹, Luyao Guan¹, Zhuo Liu³ (¹Beijing Univ. of Posts and Telecommunications (BUPT), China, ²China State Shipbuilding Corporation, China, ³China Mobile Communications Corporation, China)

Keywords:Data analytics for network control and management in optical core/data center networks , Availability/reliability/disaster recovery in big data-oriented computing systems

We propose an adaptive failure prediction scheme based on LSTM leveraging actual datasets from network operator, the experimental results verify that the average accuracy of failure prediction is 93.8%.