Efficient distribution of Service requests between fog and cloud nodes considering user mobility and fog nodes’ overload is an important issue of fog computing. This paper proposes a heuristic method for task placement considering the mobility of users, aiming to serve a higher number of requested Services and minimize their response time. This method introduces a formula to overload prediction based on the entry-exit ratio of users and the estimated time required to perform current requests that are waiting in the queue of a fog node. Then, it provides a solution to avoid the predicted overloading of fog nodes by sending all delay-tolerant requests in the overloaded fog node’s queue to the cloud to reduce the time required for servicing delay-sensitive requests and to increase their acceptance rate. In addition, to prevent requests from being rejected when the mobile user leaves the coverage area of the current fog node, the requests in the current fog node’s queue will be transferred to the destination fog node. Simulation results indicate that the proposed method is effective in avoiding the overloading of the fog nodes and outperforms the existing methods in terms of response time and acceptance rate.

Recent advances in the context of Internet of Things (IoT) have led to the emergence of many useful IoT applications with different Quality of Service (QoS) requirements. The fog-cloud computing systems o er a promising environment to provision resources for IoT application Services. However, providing an effecient solution to Service placement problem in such systems is a critical challenge. To address this challenge, in this paper, we propose a QoS-aware Service placement policy for fog-cloud computing systems that places the most delay-sensitive application Services as closer to the clients as possible. We validate our proposed algorithm in the iFogSim simulator. Results demonstrate that our algorithm achieves significant improvement in terms of Service latency and execution cost compared to simulators built-in poli-cies.

The drive of this investigation is to diagnose the mediation effect of placement between program quality, industrial link, student satisfaction, and Service quality in the circumstance of tertiary education in Malaysia. Using the Cue Utilization theory, the proposed model is tested employing data collected from 173 international students who are pursuing study at University Utara Malaysia through the convenient sampling approach. The analysis is conducted through the Structural Equation Modeling (SEM). The goodness of fit model is ensured through RMSEA (< 0. 08), CFI (> 0. 90), and Chisq/Df ratio (< 5. 0) fit indices. The findings reveal that placement mediates fully between industrial link and Service quality of education. However, placement does not confirm any significant mediation effects between program quality, student satisfaction and Service quality. Results give the impression to suggest that program quality and student satisfaction are influencing Service quality directly and significantly. The findings of this study have potentially valuable implications for Service quality improvement in higher education context of Malaysia by reinforcing placement, program quality, industrial link and student satisfaction. The results signify amply by approving the mediating influence of placement about industrial link and Service quality of education respecting international students.

Software-defined networks (SDN) have emerged as a new paradigm to overcome rigidity in traditional networks. SDN controllers manage network switches through a centralized control plane. Strategically placing controllers is vital for meeting performance needs. We model the NP-hard controller placement problem (CPP) as multi-objective optimization reconciling switch-controller latency, resilience to failures, inter-controller coordination overhead and load balancing. A customized Non-dominated Sorting Moth Flame algorithm (NS-MF) with novel recombination and perturbation techniques is proposed to effectively approximate the Pareto-optimal set of placements on large problem instances. NS-MF is benchmarked on a diverse corpus of 41 topologies against the exhaustive POCO solver, assessing computational time and solution quality tradeoffs. Compared to POCO, the proposed algorithm attains over 20X speedup for the largest graphs with an average optimality gap within 0.8%. The proposed NS-MF demonstrates superior performance over state-of-the-art metaheuristics (NSGA-II and PSA) in reconciling proximity and diversity objectives when estimating Pareto-optimal fronts. Experimental results substantiate NS-MF's efficacy in effectively navigating objectives pertinent to resilient SDN design.

Network function virtualization technology transforms hardware middleboxes into sets of software-based Virtual Network Function (VNF ) that can host the growing demand for latency-sensitive Services at the fog-cloud computing-based networks. Dynamic placement of Service functions chain by reusing (VNF) instances can improve resource utilization and save time. To address this problem, we propose a parallelized Service function chain placement method by reusing (VNF) in fog-cloud computing-based networks. Here, the placement problem is configured using deep reinforcement learning approaches with the aim of maximizing long-term cumulative reward. By sharing (VNF) s in parallel, this method can achieve computational acceleration in providing online Services. In addition, the proposed method increases the ability to accept future requests by extracting the distribution of the initialized (VNF) s. The simulation results show the superiority of the proposed method, where considering the monetary cost criterion of more than 7%, it performs better than the best existing methods.