Fog computing has emerged as a pivotal paradigm for enabling low-latency, high-performance applications by positioning computational resources closer to the network edge. However, task offloading in fog environments poses significant challenges because of the dynamic and heterogeneous nature of fog nodes, which are influenced by fluctuating computational loads, channel conditions, and mobility patterns. This paper introduces a dynamic task-offloading framework leveraging deep Q-learning (DQL), a reinforcement learning technique tailored to optimize task allocation and enhance system performance in such complex settings. The proposed framework models the offloading problem as a Markov decision process (MDP), enabling the DQL agent to learn optimal strategies adaptively by incorporating factors such as task demands, node states, and channel quality. Performance evaluations against state-of-the-art scheduling methods reveal that the DQL-based approach consistently outperforms competing techniques, achieving superior efficiency and reliability. Furthermore, the framework demonstrates scalability and robustness in dynamic fog networks, making it highly suitable for diverse real-time applications. This study highlights the potential of DQL as a transformative solution for dynamic task offloading in fog computing, offering efficient resource management and system stability. Its applicability to domains such as intelligent transportation systems, smart cities, and the IoT underscores its practical relevance and future impact.

One of the challenges in managing cloud computing clusters is assigning resources based on the customers’ needs. For this mechanism to work efficiently, it is imperative that there are sufficient resources reserved to maintain continuous operation, but not too much to avoid overhead costs. Additionally, to avoid the overhead of acquisition time, it is important to reserve resources sufficiently in advance. This paper presents a novel reliable general-purpose mechanism for prediction-based resource usage reservation. The proposed solution should be capable of operating for long periods of time without drift-related problems, and dynamically adapt to changes in system usage. To achieve this, a novel signature-based ensemble prediction method is presented, which utilizes multiple distinct prediction algorithms suited for various use-cases, as well as an anomaly detection mechanism used to improve prediction accuracy. This ensures that the mechanism can operate efficiently in different real-life scenarios. Thanks to a novel signature-based selection algorithm, it is possible to use the best available prediction algorithm for each use-case, even over long periods of time, which would typically lead to drifts. The proposed approach has been evaluated using real-life historical data from various production servers, which include traces from more than 1,500 machines collected over more than a year. Experimental results have demonstrated an increase in prediction accuracy of up to 21.4 percent over the neural network approach. The evaluation of the proposed approach highlights the importance of choosing the appropriate prediction method, especially in diverse scenarios where the load changes frequently.