Modeling Website Workload Using Neural Networks

TL;DR

This paper employs artificial neural networks to model and forecast website request loads during the 1998 FIFA World Cup, demonstrating the effectiveness of time-series forecasting with neural networks.

Contribution

It introduces a neural network-based approach for modeling website workload using time-series data from real traffic logs, with detailed data processing and experimental analysis.

Findings

Neural networks can effectively model website request patterns.

Multiple modeling cases show varying prediction accuracies.

The approach provides insights into website traffic dynamics.

Abstract

In this article, artificial neural networks (ANN) are used for modeling the number of requests received by 1998 FIFA World Cup website. Modeling is done by means of time-series forecasting. The log traces of the website, available through the Internet Traffic Archive (ITA), are processed to obtain two time-series data sets that are used for finding the following measurements: requests/day and requests/second. These are modeled by training and simulating ANN. The method followed to collect and process the data, and perform the experiments have been detailed in this article. In total, 13 cases have been tried and their results have been presented, discussed, compared and summarized. Lastly, future works have also been mentioned.