Fuzzy logic system balances fog computing loads to cut delay and energy

Fog computing sits close to the devices that generate data, extending computation and storage from the cloud into the edge network of end devices. By processing requests nearer their source, it tackles latency issues and enhances the quality of service that end-users experience. Yet this proximity creates a new problem: as the number of requests from end devices grows, the fog nodes that serve them can become overloaded, degrading the very performance the architecture was designed to protect.

Load balancing has therefore attracted significant attention as a way to enhance system throughput, latency, and energy efficiency in fog environments. Numerous research efforts have been undertaken, but the authors of a new study identify a persistent gap. Determining the load of fog nodes and migrating tasks away from overloaded ones, while aiming to minimize both delay and load, remains under-addressed in the existing literature.

Writing in the journal Cluster Computing, Nidhi Kumari and Prasanta K. Jana propose a fuzzy logic-based system to calculate the load of fog nodes. Their approach jointly considers three factors: CPU utilization, memory usage, and task queue length. Combining these signals, they argue, provides a more adaptive and accurate load estimation compared to conventional threshold-based techniques.

Estimating load, however, only identifies which nodes are strained. The second half of the proposed scheme addresses what to do about it. The researchers present an efficient and novel particle encoding scheme along with a multi-objective fitness function to migrate tasks from overloaded fog nodes to others. The researchers present this encoding scheme along with a multi-objective fitness function to migrate tasks from overloaded fog nodes to others.

The researchers evaluated their scheme through simulation against baseline algorithms. The simulation results indicate that the proposed scheme outperforms the baselines in terms of delay, load, and energy consumption. According to the simulation results, the proposed scheme outperformed the baselines in terms of delay, load, and energy consumption.

The superiority over the baseline algorithms is statistically validated for delay using the Friedman test.

The study appears in Volume 29 of Cluster Computing as article number 924, published on 31 October 2026. Its contribution lies in pairing a fuzzy logic-based load calculation with a novel particle encoding scheme and a multi-objective fitness function, then demonstrating through simulation that the combination outperforms baseline algorithms in terms of delay, load, and energy consumption, with the delay advantage statistically validated using the Friedman test.

Subject of Research: A fuzzy logic-based load balancing and task offloading approach for fog computing that estimates fog node load and migrates tasks to reduce delay, load, and energy consumption.

Article Title: Fuzzy logic-based load balancing approach for task offloading in fog computing

Article References: Kumari, N., & Jana, P. K. (2026). Fuzzy logic-based load balancing approach for task offloading in fog computing. Cluster Computing, 29(16), Article 924. https://doi.org/10.1007/s10586-026-06676-8

Image Credits: AI Generated

DOI: 10.1007/s10586-026-06676-8

Keywords: fog computing, load balancing, task offloading, fuzzy logic, CPU utilization, memory usage, task queue length, particle encoding, multi-objective fitness function, Friedman test, delay, energy consumption