ABSTRACT This study introduces a novel approach for an Admission Control Manager (ACM) for allocating users requests in Fog‐integrated Cloud (FiC), based on available physical resources while ensuring Quality of Service (QoS) and Quality of Experience (QoE). The proposed ACM leverages a hybrid model combining the Genetic Algorithm (GA) and Adaptive Neuro‐Fuzzy Inference System (ANFIS), referred to as GA‐ANFIS. The GA‐ANFIS model operates in two distinct phases to address the resource provisioning challenges of the extended three‐layer FiC architecture. In the first phase, GA is employed to optimize the initial parameters of the ANFIS, ensuring better learning and convergence. In the second phase, the optimized ANFIS model processes user request parameters for job classification to decide the FiC layers for processing. The model's effectiveness is evaluated using simulations on Google trace datasets, with performance assessed via metrics such as accuracy, execution time, and convergence rate. The results demonstrate significant improvements, including a 12.63% in accuracy and a 21.66% reduction in execution time compared to state‐of‐the‐art models. These findings establish the potential of the GA‐ANFIS model as an efficient ACM to address resource provisioning challenges in FiC.

Summary The rapid expansion of IoT systems has caused network congestion and delays in task placement and resource provisioning as usually the tasks are executed at a far location in the cloud. Fog computing reduces the computing burden of cloud data centers as well as the communication burden of the internet as fog resources are placed near the data generation points. Within Fog computing, an important challenge is the optimal task placement which is an NP‐class problem. This work applies machine learning for task clustering and addresses the task placement problem in a fog computing environment using a hybrid of two recent metaheuristics; Jaya and gray wolf optimization (GWO). The hybrid method considers optimizing the total number of active fog nodes, load balancing in fog nodes, and average response time of the tasks. The performance of the proposed method is evaluated on a real‐time LCG dataset and is compared with reinforcement learning fog scheduling (RLFS), genetic algorithm (GA), dynamic resource allocation mechanism (DRAM), load balancing and scheduling algorithm (LBSSA), and particle swarm optimization with simulated annealing (PSO‐SA) algorithms. The results demonstrate the superiority of the suggested method over the baseline techniques in terms of average improvement of 51.04% in load balance variance, 30.25% in average response time, 24.16% in execution time, and 47.10% in the number of devices used.

Summary Fog computing, an emerging technology, extends Cloud computing services to the network's edge in the proximity of the application request. This extension yields improvement in Bandwidth (BW) utilization, faster responses to Real‐Time (RT) and Internet of Things (IoT) requests, and the provision of the heterogeneous resource services. While extensive work has been conducted on resource allocation for RT and Non‐Real‐Time (NRT) requests separately in Fog as well as Cloud computing, there is limited focus on resource provisioning for mixed RT and NRT requests in the Fog‐integrated Cloud (FiC) environment. Moreover, the majority of the existing provisioning methods primarily consider parameters from the system's perspective, overlooking crucial user aspects such as deadline and request size. To address the gap, this work introduces a resource provisioning method named “Parameter Matching of Realtime and Non‐Realtime Applications (PMRNA),” which considers user parameters and resource information, gathered by the broker. The performance evaluation of the proposed model is done in CloudSim, using various combinations of RT and NRT requests along with diverse Fog and Cloud resource configurations. Evaluation metric includes average Execution Time (ET), average Waiting Time (WT), average Turn Around Time (TAT), and resource utilization. The experimental results demonstrate a significant reduction in both average WT and average TAT for the diverse pool of RT and NRT requests in the FiC compared to the Cloud‐only environment.

Abstract In a hierarchical edge‐fog‐cloud architecture, edge devices possess limited resources and energy. To contain with, it can offload some tasks generated by the Internet of Things (IoT) to the fog and cloud. Several factors influence this task‐offloading decision, including hardware features, network conditions, and application characteristics. Most of the research studies, in task offloading systems, have confined to changing parameter values, whereas very few have considered fuzzy‐based dynamic approaches for resource allocation. This work proposes a fuzzy‐based task offloading technique (FBOT) for scheduling the tasks (both compute and data‐intensive) to the appropriate nodes of the fog‐cloud system. The proposed method incorporates a vague logic‐based fuzzy task scheduler for scheduling various tasks at the fog layer. This helps to reduce the waiting time for the compute‐intensive tasks and minimizes the starvation problem of data‐intensive tasks. In addition, the proposed technique decreases the task completion time and selects the best computing nodes for each tasks. To validate the performance of the proposed technique, extensive simulation studies have been carried out on various parameters. Compared with the baseline algorithms, the results show that the proposed method offers an average improvement of 30% in terms of the total time of the task.

Summary Internet of Things (IoT) and other smart devices produce data that are large in volume, variety, and velocity. Cloud not only helps in data analysis, but also provides storage and computation facility to these data. It has been experienced that for many time‐critical applications, by the time request traverses back and forth to the cloud for analysis/execution, the opportunity to act upon it may get over. Therefore, time sensitivity and priority for such applications greatly matter. Fog computing, an upcoming computing infrastructure, complements the cloud and overcomes this limitation by supporting time‐sensitive and priority‐based applications by provisioning the computation, bandwidth, and storage. However, adopting fog‐integrated cloud introduces newer resource management challenges requiring a new request scheduling scheme with appropriate quality of service/experience (QoS/QoE). In this work, an intelligent admission control manager is being proposed for placing the request based on the parameters such as CPU, memory, storage besides few other categorical parameters, for example, job priority and time sensitivity. The proposed work applies machine intelligence techniques, clustering for labeling the applications' requests followed by a decision tree, using the labeled requests, to classify the incoming requests. The proposed methodology is demonstrated in terms of accuracy, execution time, precision, recall, variation in accuracy, and execution time by introducing noise in multiple size batches to avoid the generalization error and fault tolerance. A comparative study with few well‐known classifiers has also been performed to ascertain the effectiveness of the proposed model. The proposed model is light enough to be placed appropriately on the fog node.

Due to emergence of Internet of Things (IoT), fog computing is gaining momentum in the IT industry. The fog nodes are owned by the fog service providers (FSPs) and usually are not intended to provide their services for free. If FSP charges high for its services or does not stick with its promised quality of service, existing users of that FSP may leave early or may churn to some other FSP. In such competitive scenario, to survive and to maximize the profit in the long run, FSPs should accept the requests of the users considering both technical and non-technical parameters. Since both FSP and IoT users are strategic decision makers, game theoretic analysis may help FSPs to maximize their payoffs. With the change in the strategy of the player, equilibrium solution may change and therefore this dynamic scenario is formulated as an extensive game form. A subgame perfect equilibrium, obtained for this game using backward induction, makes admission control policies suitable for different environment which helps the FSPs in maximizing their profit in the long run. A comparative analysis of the proposed work with state of art indicates that the proposed work outperforms and generates better revenue to the FSPs.

Load-balanced scheduling deals with the uniform allocation of workload to a set of computational resources in order to optimize some characteristic metrics such as makespan, resource utilization and relative load imbalance. As such, the load balancing is an NP-hard problem which becomes more complex when heterogeneity in the workload and the computing resources are introduced. Workload heterogeneity is defined by the types of the workload whereas CPU resources can be heterogeneous in terms of memory/cache hierarchy, clock speed, etc. For load balancing in a truly heterogeneous multicore system, this work proposes a model by incorporating a related heuristic into Genetic Algorithm (GA) to generate the priorities of the workload and the computing resources by exploiting their heterogeneity characteristics. The priorities play a significant role in the effective mapping of the workload on the computing resources. A good number of simulation experiments are carried out to study the performance of the proposed model besides comparing it with some contemporary GA-based models. Results indicate that incorporating the relevant heuristic into GA makes a significant load-balanced scheduling, especially for heavy workload applications.

Resource procurement using reverse auction in Cloud computing is an interesting but a complex problem as it involves many attributes and constraints. Reverse auction is a mechanism in which a customer prepares call for proposal of resource requirement and publicizes it in order to get the attention of eligible service providers. This work proposes a multi-attribute combinatorial reverse auction for Cloud resource procurement which considers price as well as non-price attributes such as quality of service parameters, reputation etc. in the determination of winning service providers. For this, the problem is formulated using approximation algorithm and near optimal solution is obtained in polynomial time. Auction mechanism allows providers to reveal true information in order to maximize their profit. It also imposes a penalty on the providers who cheat i.e., do not offer the agreed upon services. This makes the system robust as it maintains the utility of the customer. It also maintains a healthy competition among the providers. Performance evaluation and a comparative study with some base line models exhibit that the proposed method performs better.

In Virtualized data center for implementing cloud, mapping of virtual machines (VM) on to physical machines aims to efficiently utilize the resources of the physical machines in order to earn better profit for the cloud provider. Energy is an important parameter for the huge virtualized data center and is to be minimized as much as possible while VM placement. The proposed work addresses two important Cloud aspects: efficient energy usage and effective resource utilization. For this, modified binary particle swarm optimization is proposed and applied for the VM placement problem. The simulation experiments, carried out to evaluate the performance of the proposed model, prove that the proposed approach results in reduced energy consumption with least resource wastage compared with some other contemporary methods proposed in the literature.

Cloud computing, a distributed computing paradigm, enables delivery of IT resources over the Internet and follows the pay-as-you-go billing model. Workflow scheduling is one of the most challenging problems in cloud computing. Although, workflow scheduling on distributed systems like grids and clusters have been extensively studied, however, these solutions are not viable for a cloud environment. It is because, a cloud environment differs from other distributed environment in two major ways: on-demand resource provisioning and pay-as-you-go pricing model. Thus, to achieve the true benefits of workflow orchestration onto cloud resources novel approaches that can capitalize the advantages and address the challenges specific to a cloud environment needs to be developed. This work proposes a dynamic cost-effective deadline-constrained heuristic algorithm for scheduling a scientific workflow in a public cloud. The proposed technique aims to exploit the advantages offered by cloud computing while taking into account the virtual machine (VM) performance variability and instance acquisition delay to identify a just-in-time schedule of a deadline constrained scientific workflow at lesser costs. Performance evaluation on some well-known scientific workflows exhibit that the proposed algorithm delivers better performance in comparison to the current state-of-the-art heuristics.

Summary The growing popularity of cloud services has attracted large number of business organizations to enter into the cloud market. The varied availability of cloud service providers, for various services, leaves the cloud customers into skepticism. Service measurement index (SMI) framework, which is based on the analytic hierarchical process, is developed for the ranking of cloud services. As there are multiple dependencies among the criteria for the selection of cloud service provider, analytic hierarchical process formulation is not efficient. In the proposed work, analytic network process that captures the cloud service‐provider selection problem in a more realistic way is integrated in the ranking component of SMI framework. In this method, interactions among the criteria are used for the ranking of the cloud services. Few new criterions are also added into the existing SMI framework to make the framework more effective. A case study is conducted to analyze the proposed technique for the cloud provider selection.

Summary Cloud computing is an upcoming and promising solution for utility computing that provides resources on demand. As it has grown into a business model, a large number of cloud service providers exist today in the cloud market, which further is expanding exponentially. Many cloud service providers, with almost similar functionality, pose a selection problem to the cloud users. To assist the users in the best service selection, as per its requirement, a framework has been developed in which users list their quality of service (QoS) expectation, while service providers express their offerings. Experience of the existing cloud users is also taken into account in order to select the best cloud service provider. This work identifies some new QoS metrics, besides few existing ones, and defines it in a way that eases both the user and the provider to express their expectations and offers, respectively, in a quantified manner. Further, a dynamic and flexible model, using a variant of ranked voting method, is proposed that considers users' requirement and suggests the best cloud service provider. Case studies affirm the correctness and the effectiveness of the proposed model. Copyright © 2016 John Wiley & Sons, Ltd.

A computational grid helps in faster execution for compute-intensive jobs. Many characteristic parameters are intended to be optimized while making resource allocation for job execution in a computational grid. Most often, the green energy aspect, where one tries for better energy utilization, is ignored while allocating grid resources to jobs. The present work tries to optimize energy to make it a green energy model for resource allocation in a computational grid. It explores how effectively the jobs submitted to the grid can be executed, making no compromise on other expected characteristic parameters, and at the same time makes optimal use of energy. The proposed green energy model has been experimentally verified by its simulation. The result reveals the benefits and gives a better insight for effective resource allocation.

SUMMARY Computational grid provides a wide distributed platform for high‐end compute intensive applications. Grid scheduling is often carried out to schedule the submitted jobs on the nodes of the grid so that some characteristic parameter is optimized. Availability of the computational nodes is one of the important characteristic parameters and measures the probability of the node availability for job execution. This paper addresses the availability of the grid computational nodes for the job execution and proposes a model to maximize it. As such, the task scheduling problem in grid is nondeterministic polynomial‐time hard, and often, metaheuristics techniques are applied to solve it. Genetic algorithm, a metaheuristic technique based on evolutionary computation, has been used to solve such complex optimization problem. This work proposes a technique for the grid scheduling problem using genetic algorithm with the objective to maximize availability. Simulation experiment, to evaluate the performance of the proposed algorithm, is conducted, and results reveal the effectiveness of the model. A comparative study has also been performed. Copyright © 2014 John Wiley & Sons, Ltd.

SUMMARY Computational grids allow the sharing of geographically distributed computational resources in an efficient, reliable, and secure manner. Grid is still in its infancy, and there are many problems associated with the computational grid, namely job scheduling, resource management, information service, information security, routing, fault tolerance, and many more. Scheduling of jobs on grid nodes is an NP‐class problem warranting for heuristic and meta‐heuristic solution approach. In the proposed work, a meta‐heuristic technique, auto controlled ant colony optimization, has been applied to solve this problem. The work observes the effect of interprocess communication in process to optimize turnaround time of the job. The proposed model has been simulated in Matlab. For the different scenarios in computational grid, results have been analyzed. Result of the proposed model is compared with another meta‐heuristic technique genetic algorithm that has been applied for the same purpose. It is found that auto controlled ant colony optimization not only gives better solution in comparison to genetic algorithm, but also converges faster because initial solution itself is good because of constructive and decision‐based policy adapted by the former. Concurrency and Computation: Practice and Experience, 2012.© 2012 Wiley Periodicals, Inc.

SUMMARY Grid applications with stringent security requirements introduce challenging concerns because the schedule devised by nonsecurity‐aware scheduling algorithms may suffer in scheduling security constraints tasks. To make security‐aware scheduling, estimation and quantification of security overhead is necessary. The proposed model quantifies security, in the form of security levels, on the basis of the negotiated cipher suite between task and the grid‐node and incorporates it into existing heuristics MinMin and MaxMin to make it security‐aware MinMin(SA) and MaxMin(SA). It also proposes SPMaxMin (Security Prioritized MinMin) and its comparison with three heuristics MinMin(SA), MaxMin(SA), and SPMinMin on heterogeneous grid/task environment. Extensive computer simulation results reveal that the performance of the various heuristics varies with the variation in computational and security heterogeneity. Its analysis over nine heterogeneous grid/task workload situations indicates that an algorithm that performs better for one workload degrades in another. It is conspicuous that for a particular workload one algorithm gives better makespan while another gives better response time. Finally, a security‐aware scheduling model is proposed, which adapts itself to the dynamic nature of the grid and picks the best suited algorithm among the four analyzed heuristics on the basis of job characteristics, grid characteristics, and desired performance metric. Copyright © 2011 John Wiley & Sons, Ltd.

This paper exploits the potential of the Genetic Algorithm to solve the cellular resource allocation problem. When a blocked host is to be allocated to a borrowable channel, a crucial decision is which neighboring cell to choose to borrow a channel. It is an optimization problem and the genetic algorithm is efficiently applied to handle this. The Genetic Algorithm, for this particular problem, is improved by introducing a new genetic operator, named pluck, that incorporates a problem-specific knowledge in population generation and leads to a better channel utilization by reducing the average blocked hosts. The pluck operator makes the crucial decision of when and which cell to borrow with the future consideration that the borrowing should not lead the network to chaos. It makes a channel borrowing decision that minimizes the number of blocked hosts and improves the long-term performance of the network. Efficacy of the proposed method has been evaluated by experimentation.

Summary form only given. Most of the task allocation models & algorithms in distributed computing system (DCS) require a priori knowledge of its execution time on the processing nodes. Since the task assignment is not known in advance, this time is quite difficult to estimate. We propose a cluster-based dynamic allocation scheme, in a distributed computing system, which eliminate this time requirement. Further, as opposed to a single task allocation, generally proposed in most of the models, we consider multiple tasks. A fuzzy function is used for both the module clustering and processor clustering. Dynamic invocation of clustering and assignment is considered. Experimental results show the efficacy of the proposed model.