A comparative analysis of network modelling approaches for predicting performance in large scale networks
Singh, Khushmeet Agarwal, Pratik Rana, Vikrant Kolli, Jayanth Cheruku, Saketh R. Musunuri, Aravindsundeep
Singh, Khushmeet
Agarwal, Pratik
Rana, Vikrant
Kolli, Jayanth
Cheruku, Saketh R.
Musunuri, Aravindsundeep
Other Names
Location
Time Period
Advisors
Original Date
Digitization Date
Issue Date
2025-04-01
Type
Conference paper
Genre
Keywords
Crucial,Detecting,Modeling,Network,Numerous,Performance,Prediction,Requirement
Subjects (LCSH)
Citation
K. Singh, P. Agarwal, V. Rana, J. Kolli, S. R. Cheruku and A. Musunuri, "A Comparative analysis of Network Modelling Approaches for Predicting Performance in Large Scale Networks," 2025 International Conference on Pervasive Computational Technologies (ICPCT), Greater Noida, India, 2025, pp. 650-655, doi: 10.1109/ICPCT64145.2025.10939104.
Abstract
choosing the right node to measure is the key to success in large-scale networks; network modeling is a critical tool for detecting performance. The growing size and complexity of modern networks make accurately predicting their performance a crucial requirement for efficient and reliable operation. Numerous network modeling approaches, however, each with unique strengths and limitations, have emerged in response to this. This paper provides a comparative critique of these network modeling approaches and their performance prediction capabilities in large-scale networks. The first category is the analytical modeling approach, where mathematical equations represent network performance. An approach similar to the described process gives very accurate and detailed predictions but must rely on a theoretical understanding of how networks behave and the assumption of ideal networks. The second approach, simulation modeling, designs a virtual model of the network and lets you test it to predict its performance. This allows for greater flexibility, and real-world factors can be included in the heuristics; however, it is potentially very computationally intensive. The third method is machine learning, which involves algorithms that learn patterns in historical network data and create predictions. While this method is able to process intricate data sets and adjust to evolving network states, its precision relies heavily on the quality of the training data. © 2025 IEEE.
Table of Contents
Description
Click on the DOI link to access this article at the publishers website (may not be free).
Publisher
Institute of Electrical and Electronics Engineers Inc.
Journal
Book Title
Series
2025 International Conference on Pervasive Computational Technologies, ICPCT 2025
8 February 2025 through 9 February 2025
Greater Noida
208025
8 February 2025 through 9 February 2025
Greater Noida
208025