Abstract

SEGARRA-GUZMAN, Edison  and  LUDENA-GONZALEZ, Patricia. Distributed Congestion Control Based on Utility Function. Enfoque UTE [online]. 2024, vol.15, n.2, pp.9-19. ISSN 1390-6542.  https://doi.org/10.29019/enfoqueute.994.

This paper introduces the Distributed Utility Function Algorithm (D-AFU) as a notable progression in managing and optimizing network traffic within distributed settings. Based on the utility function principle, D-AFU dynamically adjusts data rate in response to ever-changing network demands, with optimal performance and a higher user experience. Contrary to the centralized model, D-AFU employs a distributed, scalable and resilient against failures and system overloads mechanism. Its efficiency was validated using the NS-3 simulator. Three main metrics were used: the data rate allocation, utility per session, and fairness (quantified by the Gini coefficient). D-AFU displays exceptional performance and low latency, particularly vital for real-time applications with high Quality of Service (QoS) requirements.

Keywords : Congestion Control; Utility Function; Real-Time applications; Elastic Applications; Distributed Optimization; Proactive Algorithm.