Comparative Analysis of Deep Learning Models for Real-World ISP Network Traffic Forecasting
By: Josef Koumar , Timotej Smoleň , Kamil Jeřábek and more
Potential Business Impact:
Helps internet providers predict traffic jams.
Accurate network traffic forecasting is essential for Internet Service Providers (ISP) to optimize resources, enhance user experience, and mitigate anomalies. This study evaluates state-of-the-art deep learning models on CESNET-TimeSeries24, a recently published, comprehensive real-world network traffic dataset from the ISP network CESNET3 spanning multivariate time series over 40 weeks. Our findings highlight the balance between prediction accuracy and computational efficiency across different levels of network granularity. Additionally, this work establishes a reproducible methodology that facilitates direct comparison of existing approaches, explores their strengths and weaknesses, and provides a benchmark for future studies using this dataset.
Similar Papers
Network Traffic Classification Using Machine Learning, Transformer, and Large Language Models
Machine Learning (CS)
Sorts internet traffic into types accurately.
A comparative study of deep learning and ensemble learning to extend the horizon of traffic forecasting
Machine Learning (CS)
Predicts traffic jams up to 30 days ahead.
Improving Internet Traffic Matrix Prediction via Time Series Clustering
Machine Learning (CS)
Makes internet faster by predicting traffic jams.