Convolutional LSTM models to estimate network traffic

Abstract

Network utilisation efficiency can, at least in principle, often be improved by dynamically re-configuring routing policies to better distribute ongoing large data transfers. Unfortunately, the information necessary to decide on an appropriate reconfiguration—details of on-going and upcoming data transfers such as their source and destination and, most importantly, their volume and duration—is usually lacking. Fortunately, the increased use of scheduled transfer services, such as FTS, makes it possible to collect the necessary information. However, the mere detection and characterisation of larger transfers is not sufficient to predict with confidence the likelihood a network link will become overloaded. In this paper we present the use of LSTM-based models (CNN-LSTM and Conv-LSTM) to effiectively estimate future network traffic and so provide a solid basis for formulating a sensible network configuration plan.

NOTED: An intelligent network controller to improve the throughput of large data transfers in File Transfer Services by handling dynamic circuits

“…First achievements and outcomes are documented in articles [2,3]. Furthermore, a comprehensive study on traffic forecasting has been conducted using a machine learning approach with Long Short Term Memory (LSTM) neural networks, as detailed in [4]. In terms of traffic engineering, the implementation of NOTED presents certain challenges that need to be addressed.…”