Downloads: 0 | Views: 25
Analysis Study Research Paper | Information Technology | Kenya | Volume 14 Issue 4, April 2025 | Rating: 4.6 / 10
Ranking and Validation of Variables Determining Fog Computing Performance in IoT Applications
Kanuku Watson, Njenga Stephen, Musumba George
Abstract: Cloud computing capabilities are extended to the network's edge with fog computing, which offers low-latency processing and storage, making it essential for Internet of Things (IoT) applications. Numerous network and computational factors, however, can significantly influence a substantial impact on fog computing systems' performance. The main objective of this study is to rank and validate the most important elements that affect fog computing performance, particularly when it comes to Internet of Things applications. We investigate five key variables: Packet Loss Rate (PLR), Queue Time, Latency, Channel Utilization, and Throughput, in terms of their influence on the overall performance of fog nodes and IoT systems. Through a comprehensive evaluation using both theoretical models and experimental data, we establish a ranking for these variables based on their direct impact on fog computing performance. The analysis shows that Packet Loss Rate emerges as the most critical factor, as higher packet loss can severely degrade the reliability of communication between fog nodes and IoT devices. This is followed by Queue Time, which represents the delay incurred in processing incoming data requests; longer queue times contribute to increased system latencies and reduced throughput. Latency itself, although related to the aforementioned factors, ranks third as it directly affects the responsiveness of real-time applications. Channel Utilization, a measure of how effectively the communication channel is used, ranks fourth, influencing the overall network capacity and bandwidth efficiency. Lastly, Throughput is ranked fifth, as it is closely tied to the network's ability to transmit data efficiently but has a secondary effect compared to other variables in terms of performance degradation. These rankings were validated through consultation with industry experts, confirming the crucial roles that PLR and Queue Time play in optimizing fog computing performance. These findings provide valuable insights for researchers and practitioners seeking to improve the design and implementation of fog computing systems, highlighting the need for targeted optimizations in the most impactful variables to achieve enhanced performance in IoT environments.
Keywords: Fog computing, IoT, Packet Loss Rate, Queue Time, Latency, Channel Utilization, Throughput, Performance ranking, Network optimization
Edition: Volume 14 Issue 4, April 2025,
Pages: 1057 - 1063