Exécution de notre outil de synthèse sur un nouvel article
Vérifiez l’article ci-dessous, puis lancez sa synthèse.
Short-Term Internet Network Traffic Patterns in Greece: A High-Frequency Time-Series Analysis
Constantinos Challoumis - Κωνσταντίνος Χαλλουμής (National and Kapodistrian University of Athens, University of West Attica), Konstantinos Mavrommatis (University of West Attica)
Résumé : This study examines short-term Internet traffic patterns in Greece within the context of time-series dynamics and digital infrastructure behavior, aiming to identify temporal structures, daily cycles, and variability in network traffic. Using high-frequency data collected at 5-minute intervals over three months, this study adopts a multi-scale time-series analysis approach to investigate short-term Internet traffic dynamics. The analysis aggregates traffic data at hourly and daily levels and applies descriptive statistical analyses and visualization techniques, including hourly aggregation, daily averages, boxplots, and heatmaps. The findings reveal a strong and consistent diurnal pattern, with minimum traffic during the early morning hours (02:00–05:00) and peak demand in the evening (18:00–21:00). Inbound and outbound traffic exhibit near-perfect symmetry, indicating balanced network flows. Although mean traffic remains stable, significant short-term fluctuations and increased variability during peak periods are observed, highlighting both the stability and the dynamic nature of network behavior.