Abstract Fog computing emerges as the new computing environment that stays in the proximity of end‐users and harnesses resources at the edge of the network to extend cloud‐facilities. It provides attractive solutions to the diverse range of Internet of Things (IoT) applications by executing them in the vicinity of end‐users. It is challenging to schedule these latency‐sensitive, computation‐intensive, and resource‐hungry applications on distributed, heterogeneous, and resource‐constrained Fog computing environment while ensuring time‐bound service delivery and satisfying Quality of Service (QoS) requirements of end‐users. In this paper, a context‐aware application scheduling technique is proposed for Fog computing environments that employs various parameters of device‐ and application‐level context to minimize service delivery time and satisfy QoS requirements of various IoT applications such as surveillance and game‐based applications. The performance of the proposed technique is evaluated in a simulated Fog environment and compared with baseline application scheduling techniques. The simulation results demonstrate that the proposed context‐aware scheduling techniques result in significant improvement in service delivery time and QoS compared to baseline scheduling techniques.

Summary Efficient utilization of resources is a challenging task in cloud computing due to heterogeneity of resources and fluctuating resource demands of multifarious applications. This article presents an autonomic resource allocation approach for energy efficient utilization of resources while satisfying QoS needs of the end‐users. The proposed approach is based on antlion optimization and makes use of resource utilization history for autonomic allocation and efficient use of resources. It can operate in one of the three different modes namely, energy‐saving, performance, and balanced mode depending on the projected workload of an application. The performance analysis of the proposed approach is carried out in CloudSim simulator under different workload conditions.

SUMMARY Load balancing (LB) is nothing but the systematic distribution of load over different servers. The fog server is handling the maximum data of the cloud server to enhance the advancement of users' requests. The growth in data requests is escalating, and fog computing has intensified the accessibility of the data. Fog computing achieves many challenges according to the demands of the users, but even so, some challenges require more progress. The problem faced by fog computing is LB due to an increase in traffic on the network layer. Various LB techniques have already been proposed in the cloud layer, but until now they have only been in progress in the fog layer. Inefficient LB may cause a decrease in service quality, like delays in response time, processing time, security, and many more. In this survey, several algorithms have been discussed that are based on LB, which works out the issue of overloaded data on the network. Some parameters that authors have focused in LB are latency, bandwidth, deadlines, cost, security, execution time, and response time. Other parameters based on fault tolerance are also discussed with their quality parameter table and algorithm. In addition to this some of the limitations of the author's work, that is discussed in this article.

Abstract The biggest challenge of task scheduling in Fog computing is to satisfy users' dynamic requirements in real‐time with Fog nodes' limited resource capacities. Fog nodes' heterogeneity and an obligation to complete tasks by the deadline while minimizing cost and energy consumption makes the scheduling process more challenging. This article facilitates a deeper understanding of the research issues through a detailed taxonomy and distinguishes significant challenges in existing work. Furthermore, the paper investigates existing solutions for various challenges, presents a meta‐analysis on quality of service parameters and tools used to implement Fog task scheduling algorithms. This systematic review will help potential researchers easily identify specific research problems and future directions to enhance scheduling efficiency.

Summary With the growing popularity of cloud computing, different infrastructure as a service cloud frameworks do exists. Each framework has significant impact on robustness, scalability, fault‐tolerance, energy efficiency, and so on of the cloud. To bring good characteristics of open source clouds and commercial clouds under a roof, a six layered green cloud framework namely SERVmegh is proposed. Resource management, green power management, workload analyzer and manager, and on/off control components of SERVmegh framework are designed and developed. Energy‐efficient resource wastage reduction methodology for resource management is proposed. Algorithms for resource management, virtual machine placement, and minimization of virtual machine migrations using resource wastage reduction methodology are implemented. The performance of proposed algorithms is evaluated in CloudSim simulation and OpenNubula open source cloud environment. The results depict significant energy savings under different scenarios. Copyright © 2016 John Wiley & Sons, Ltd.