In recent years, the Internet of Things (IoT) has been one of the most popular technologies that facilitate new interactions among things and humans to enhance the quality of life. With the rapid development of IoT, the fog computing paradigm is emerging as an attractive solution for processing the data of IoT applications. In the fog environment, IoT applications are executed by the intermediate computing nodes in the fog, as well as the physical servers in cloud data centers. On the other hand, due to the resource limitations, resource heterogeneity, dynamic nature, and unpredictability of fog environment, it necessitates the resource management issues as one of the challenging problems to be considered in the fog landscape. Despite the importance of resource management issues, to the best of our knowledge, there is not any systematic, comprehensive and detailed survey on the field of resource management approaches in the fog computing context. In this paper, we provide a systematic literature review (SLR) on the resource management approaches in fog environment in the form of a classical taxonomy to recognize the state-of-the-art mechanisms on this important topic and providing open issues as well. The presented taxonomy are classified into six main fields: application placement, resource scheduling, task offloading, load balancing, resource allocation, and resource provisioning. The resource management approaches are compared with each other according to the important factors such as the performance metrics, case studies, utilized techniques, and evaluation tools as well as their advantages and disadvantages are discussed.

Resource Management Approaches in Fog Computing: a Comprehensive Review

The adoption of advanced Internet of Things (IoT) technologies has impressively improved in recent years by placing such services at the extreme Edge of the network. There are, however, specific Quality of Service (QoS) trade-offs that must be considered, particularly in situations when workloads vary over time or when IoT devices are dynamically changing their geographic position. This article proposes an innovative capillary computing architecture, which benefits from mainstream Fog and Cloud computing approaches and relies on a set of new services, including an Edge/Fog/Cloud Monitoring System and a Capillary Container Orchestrator. All necessary Microservices are implemented as Docker containers, and their orchestration is performed from the Edge computing nodes up to Fog and Cloud servers in the geographic vicinity of moving IoT devices. A car equipped with a Motorhome Artificial Intelligence Communication Hardware (MACH) system as an Edge node connected to several Fog and Cloud computing servers was used for testing. Compared to using a fixed centralized Cloud provider, the service response time provided by our proposed capillary computing architecture was almost four times faster according to the 99th percentile value along with a significantly smaller standard deviation, which represents a high QoS.

https://www.mdpi.com/1424-8220/18/9/2938/pdf

A Capillary Computing Architecture for Dynamic Internet of Things: Orchestration of Microservices from Edge Devices to Fog and Cloud Providers.

Mobile cloud computing (MCC) integrates cloud computing (CC) into mobile networks, prolonging the battery life of the mobile users (MUs). However, this mode may cause significant execution delay. To address the delay issue, a new mode known as mobile edge computing (MEC) has been proposed. MEC provides computing and storage service for the edge of network, which enables MUs to execute applications efficiently and meet the delay requirements. In this paper, we present a comprehensive survey of the MEC research from the perspective of service adoption and provision. We first describe the overview of MEC, including the definition, architecture, and service of MEC. After that we review the existing MUs-oriented service adoption of MEC, i.e., offloading. More specifically, the study on offloading is divided into two key taxonomies: computation offloading and data offloading. In addition, each of them is further divided into single MU offloading scheme and multi-MU offloading scheme. Then we survey edge server- (ES-) oriented service provision, including technical indicators, ES placement, and resource allocation. In addition, other issues like applications on MEC and open issues are investigated. Finally, we conclude the paper.

/pdf/a-survey-on-mobile-edge-computing-focusing-on-service-15xvjhi6rt.pdf

A Survey on Mobile Edge Computing: Focusing on Service Adoption and Provision

Multi-dimensional quality-driven service recommendation with privacy-preservation in mobile edge environment

Mobile edge computing is becoming a promising computing architecture to overcome the resource limitation of mobile devices and bandwidth bottleneck of the core networks in mobile cloud computing. Although offloading applications to the cloud can extend the performance for the mobile devices, it may also lead to greater processing latency. Usually, the mobile users have to pay for the cloudlet resource or cloud resource they used. In this paper, we bring a thorough study on the energy consumption, time consumption, and cost of using the resource of cloudlet and cloud for workflow applications in mobile edge computing. Based on theoretical analysis, a multi-objective optimization model is established. In addition, a corresponding multi-objective computation offloading method based on non-dominated sorting genetic algorithm II is proposed to find the optimal offloading strategy for all the workflow applications. Finally, extensive experimental evaluations are performed to show that our proposed method is effective and energy- and cost-aware for workflow applications in MEC.

/pdf/an-energy-and-cost-aware-computation-offloading-method-for-4vez6b1c6s.pdf

An energy- and cost-aware computation offloading method for workflow applications in mobile edge computing

Fog computing is emerging as a powerful and popular computing paradigm to perform IoT (Internet of Things) applications, which is an extension to the cloud computing paradigm to make it possible to execute the IoT applications in the network of edge. The IoT applications could choose fog or cloud computing nodes for responding to the resource requirements, and load balancing is one of the key factors to achieve resource efficiency and avoid bottlenecks, overload, and low load. However, it is still a challenge to realize the load balance for the computing nodes in the fog environment during the execution of IoT applications. In view of this challenge, a dynamic resource allocation method, named DRAM, for load balancing in fog environment is proposed in this paper. Technically, a system framework for fog computing and the load-balance analysis for various types of computing nodes are presented first. Then, a corresponding resource allocation method in the fog environment is designed through static resource allocation and dynamic service migration to achieve the load balance for the fog computing systems. Experimental evaluation and comparison analysis are conducted to validate the efficiency and effectiveness of DRAM.

/pdf/dynamic-resource-allocation-for-load-balancing-in-fog-393u8qsakk.pdf

Dynamic Resource Allocation for Load Balancing in Fog Environment

Crowdsourcing is emerging as a powerful paradigm that utilizes the distributed devices to sense, collect, and upload data to satisfy the requirements of the users. Currently, with the popularity of edge computing, edge device users can act as recruiters or participants to publish or perform crowdsourcing tasks and share feedback. However, due to the individual selfishness, it is still a challenge to maximize the social welfare distribution of all the participants and the recruiters for the crowdsourcing market. In view of this challenge, an incentive mechanism for the crowdsourcing market with social welfare maximization in the cloud‐edge computing is designed in this paper. Technically, a double action model under the cloud‐edge computing framework is proposed first, which aims to maximize the social welfare maximization and meanwhile meet the demands of incentive compatibility, individual rationality, market clearing, and budget constraint. Secondly, a corresponding incentive mechanism is designed based on the double auction model to achieve the market fairness. Experimental evaluation and comparison analysis are conducted to validate the efficiency and effectiveness of the mechanism.

An incentive mechanism for crowdsourcing markets with social welfare maximization in cloud-edge computing

The invention discloses a virtual machine scheduling method for supporting energy consumption optimization of a cloud data center. The method comprises the following steps that firstly, a data set isrecorded based on the occupancy of a virtual machine case, and a list of physical machines in a running state in the cloud data center and a list of virtual machines in a running state in the cloud data center are obtained; secondly, according to the occupancy state of the virtual machine case, the free space of the physical machines in the running state in the first step is obtained; thirdly, according to the resource requirements of the vertical machines and the free space, obtained in the second step, of the physical machines, heuristic search is adopted, and feasible virtual machine migration strategies are obtained; fourthly, energy consumption values of the cloud data center for task execution and virtual machine migration are calculated; fifthly, according to the virtual machine migration strategies obtained in the third step, a set of the energy consumption values generated by the cloud data center after virtual machine migration operation is calculated by using the fourth step; sixthly, according to the energy consumption values obtained in the fifth step, the virtual machine migration strategy which is optimal in energy consumption is selected as the final virtual machinescheduling strategy.

Virtual machine scheduling method for supporting energy consumption optimization of cloud data center

Mobile cloud computing is emerging as a powerful platform for running demanding applications migrated from mobile devices to a remote cloud. For some real-time or urgent deadline-constrained applications, the migration process generates intolerable transmission latency. Cloudlets co-located with Access Points (APs) are considered as an efficient way to reduce such transmission latency. However, it is still a challenge to manage the cloudlets that have been deployed for fixed context-aware applications to achieve cost savings. In view of this challenge, a cloud service enhanced method supporting context-aware applications is proposed in this paper. Specifically, a cloudlet management principle is designed to provide a reference for cloudlet status judgment. Then a relevant cloud service enhanced method is proposed to decide which active cloudlets should be shut down. Finally, the experimental and analytical results demonstrate the validity of our proposed method.

/pdf/a-cloud-service-enhanced-method-supporting-context-aware-1580cycol3.pdf

A Cloud Service Enhanced Method Supporting Context-Aware Applications

Edge computing paradigms such as cloudlet, mobile edge computing (MEC) and fog computing, are promising to reduce network transmission delay while meeting intensive computing requirements at the network edge. As always assumed in theoretical studies, to expand service capability, it is desirable to integrate edge resources into a unified resource pool and make different edge facilities collaborate. However, this collaboration is difficult in practice since the heterogeneity among paradigms leads to resource isolation. This article focuses on clarifying and shielding heterogeneity among edge computing paradigms for collaboration. First, based on cloudlet, mobile edge computing and fog computing, three main edge computing paradigms, i.e., small cloudy edge paradigm (SmallC), mobile integrated edge paradigm (MobInt) and front ad hoc edge paradigm (FrontAH), are defined. So that the heterogeneity among them can be analyzed, according to which, collaboration challenges in computing resource trading, information communication and resource sharing are identified. Promising techniques such as blockchain-based smart contract, media-independent handover (MIH), and software-defined network and network function virtualization (SDN/NFV) are then introduced and integrated into a complete collaboration solution. Finally, taking the unmanned vehicular navigation as an example, the performance of collaboration is evaluated, considering additional costs introduced by collaborations, including trading, network handoff and data transmission among edge facilities in different paradigms. Simulation results show that compared with existing schemes of the three paradigms, the collaboration can reduce the cost (i.e. weighted sum of energy consumption and delay) by at least 45.6%. 

Qing Cai

Papers

Dynamic Resource Allocation for Load Balancing in Fog Environment

An incentive mechanism for crowdsourcing markets with social welfare maximization in cloud-edge computing

Virtual machine scheduling method for supporting energy consumption optimization of cloud data center

A Cloud Service Enhanced Method Supporting Context-Aware Applications

Collaboration of Heterogeneous Edge Computing Paradigms: How to Fill the Gap between Theory and Practice