In cloud computing, the resources and memory are dynamically allocated to the user based on their needs. Security is considered as a major issue in cloud as the use of cloud is increased. Intrusion detection is considered as a significant tool to develop a reliable and secure cloud environment. Performing intrusion detection in cloud is a difficult task because of its distributed nature and extensive usage. Intrusion detection system (IDS) is widely considered to find the malicious actions in network. In cloud, most conventional IDS are vulnerable to attacks and have no capability for maintaining the balance between sensitivity and accuracy. Thus, we proposed an effective dragonfly improved invasive weed optimization‐based Shepard convolutional neural network (DIIWO‐based ShCNN) to detect the intruders and to mitigate the attacks in cloud paradigm and are more feasible to detect the intruders with ShCNN. The proposed method outperforms the existing method with maximum accuracy of 0.960%, sensitivity of 0.967%, and specificity 0.961%, respectively.

A network intrusion detection system in cloud computing environment using dragonfly improved invasive weed optimization integrated Shepard convolutional neural network

Fog computing has been prioritized over cloud computing in terms of latency-sensitive Internet of Things (IoT) based services. We consider a limited resource-based fog system where real-time tasks with heterogeneous resource configurations are required to allocate within the execution deadline. Two modules are designed to handle the real-time continuous streaming tasks. The first module is task classification and buffering (TCB), which classifies the task heterogeneity using dynamic fuzzy c-means clustering and buffers into parallel virtual queues according to enhanced least laxity time. The second module is task offloading and optimal resource allocation (TOORA), which decides to offload the task either to cloud or fog and also optimally assigns the resources of fog nodes using the whale optimization algorithm, which provides high throughput. The simulation results of our proposed algorithm, called whale optimized resource allocation (WORA), is compared with results of other models, such as shortest job first (SJF), multi-objective monotone increasing sorting-based (MOMIS) algorithm, and Fuzzy Logic based Real-time Task Scheduling (FLRTS) algorithm. When 100 to 700 tasks are executed in 15 fog nodes, the results show that the WORA algorithm saves 10.3% of the average cost of MOMIS and 21.9% of the average cost of FLRTS. When comparing the energy consumption, WORA consumes 18.5% less than MOMIS and 30.8% less than FLRTS. The WORA also performed 6.4% better than MOMIS and 12.9% better than FLRTS in terms of makespan and 2.6% better than MOMIS and 4.3% better than FLRTS in terms of successful completion of tasks.

https://www.mdpi.com/2079-9292/11/19/3207/pdf?version=1665460253

A Whale Optimization Algorithm Based Resource Allocation Scheme for Cloud-Fog Based IoT Applications

The expeditious growth of the Internet of Things (IoT), ubiquitous, networked computer gadgets will pervade commercial and private areas. Devices are accessed through the Internet, and resources are kept in the cloud. Because IoT devices are power-constrained, they cannot connect to the internet or the cloud directly. Fog Computing is used to link IoT devices to the Internet or the Cloud. Fog computing is a cloud computing extension in which processing is performed on edge devices or adjacent localized data centers or cloudlets. The most crucial factors influencing fog computing performance is resource management. Due to resource constraints, resource planning is a major concern in the IoT-Cloud-Fog system. Numerous studies have introduced optimization techniques like max min , FCFS, Round robin, min min, GA. As a result of this, the IoT-Fog-Cloud System saves time, money, and energy. The goal of resource scheduling is to choose the optimal resource from a variety of matching resources which are physically available. Following scheduling, resource allocation is carried out, with the goal of allocating the chosen resource to the job. In a Cloud context, resource allocation refers to the process of assigning available virtual machine instances to workloads. This paper discusses and analyzes Min-Min, Round robin, FCFS, SJF suitable for secluding considering the following parameters average waiting time , average response time and makespan. Finally, we propose a hybrid algorithm and compare the recommended algorithm with the existing algorithm in terms of effectiveness

Optimization of Resource Scheduling and Allocation Algorithms

Internet of Things applications can determine how computing fog nodes respond to special resource requirements and distribute the load in the network. Among the most important challenges in fog-based environment of IoT applications are ensuring efficient resources allocation, avoiding bottlenecks, overloading, delays, and distributing the load among network resources. At the same time, it remains very difficult to balance the load due to computing complexity. To address these issues, this work aims to optimally distribute the network load using a hybrid approach of Fog-IoT load balancing resource allocation method with least connection and weighted round robin algorithms. It also presents an analysis of the different types of data transmission (web request, FTP file uploading/downloading). The evaluation metrics used are throughput, latency, response time, packet loss rate, channel idle, channel utilization, queuing time, speed, and required time. The results show that the proposed method achieved a total system latency of (140.05 sec), a total response time of (151.379 ms), and a total packet loss rate of (15.5%) The average total channel acquisition as idle is (99.221%), and the average speed for (256KB to 16 MB file) is 237.06 KB/sec. 

Improving Fog Computing Performance Using a Hybrid Efficient Resource Allocation Load Balancing Algorithm

The massive amounts of data generated and disbursed by digitalization necessitate computing, storage, and communication resources. The usefulness and accessibility of cloud and fog architectures becomes suitable alternatives to meet the expectations. The effectiveness with which a computing architecture's resources are allocated to service its requests determines its success. Fog computing, often refereed to as fogging fog networking is a decentralised architecture that acts an interface between the cloud and the devices that generated the data. The computing and storage requirements of the Internet of Things (IoT) can be met by using the cloud, which is an information and communication technology-based operational model. Regardless of its widespread use, cloud computing paradigm is inappropriate for all the applications based on IoT. Cloud is not suggested for applications that require low latency, such as smart transportation, health care and smart homes. This paper presents the strategies for resource distribution in fog computing along with a comparison the various techniques. 

Big Picture Analysis of Resource Allocation in Fog Computing

https://www.techscience.com/cmc/v70n2/44610/pdf

Efficient Resource Allocation in Fog Computing Using QTCS Model

https://www.techscience.com/cmc/v69n3/44128/pdf

FogQSYM: An Industry 4.0 Analytical Model for Fog Applications

Power management has turned into a basic plan parameter as more transistors are coordinated on a solitary chip. Bringing down the supply voltage is one of the alluring ways to deal with spare intensity of the variable outstanding task at hand framework, and to accomplish long battery life. DVFS is one of the effective methods to lessen the vitality utilization. The principle thought behind DVFS plan is to powerfully scale the supply voltage of CPU, to give enough circuit speed to process the framework remaining burden so as to meet the time and execution, along these lines lessening power. The problem faced by users in cloud computing is energy consumption. Some resources in cloud stay idle for some time. This idleness of resources consume energy thereby causing more consumption of energy. Both the hardware and software reduces the energy consumptions. The results shows that proposed model reduces the consumption of energy by using DVFS.

Improving Energy Consumption by Using DVFS

The recent advancement of computing paradigm named “Fog Computing” helps to satisfy the real-time latency sensitive, geo-distributed applications that requires high computational demand. Fog computing function as a middle layer in-between sensors and IoT that brings computation, storage and network functionality of cloud. This paper presents a comprehensive analysis and challenges in fog computing and provides a taxonomy of these challenges and properties with their research challenges and future research directions.

M. Iyapparaja

Papers

Efficient Resource Allocation in Fog Computing Using QTCS Model

FogQSYM: An Industry 4.0 Analytical Model for Fog Applications

Improving Energy Consumption by Using DVFS

Fog Computing: State-of-Art, Open Issues, Challenges and Future Directions