Cloud computing is a new technology which managed by a third party “cloud provider” to provide the clients with services anywhere, at any time, and under various circumstances. In order to provide clients with cloud resources and satisfy their needs, cloud computing employs virtualization and resource provisioning techniques.  The process of providing clients with shared virtualized resources (hardware, software, and platform) is a big challenge for the cloud provider because of over-provision and under-provision problems. Therefore, this paper highlighted some proposed approaches and scheduling algorithms applied for resource allocation within cloud computing through virtualization in the datacenter. The paper also aims to explore the role of virtualization in providing resources effectively based on clients’ requirements. The results of these approaches showed that each proposed approach and scheduling algorithm has an obvious role in utilizing the shared resources of the cloud data center. The paper also explored that virtualization technique has a significant impact on enhancing the network performance, save the cost by reducing the number of Physical Machines (PM) in the datacenter, balance the load, conserve the server’s energy, and allocate resources actively thus satisfying the clients’ requirements. Based on our review, the availability of Virtual Machine (VM) resource and execution time of requests are the key factors to be considered in any optimal resource allocation algorithm. As a results of our analyzing for the proposed approaches is that the requests execution time and VM availability are main issues and should in consideration in any allocating resource approach.

Intrusion detection is one of the most critical network security problems in the technology world. Machine learning techniques are being implemented to improve the Intrusion Detection System (IDS). In order to enhance the performance of IDS, different classification algorithms are applied to detect various types of attacks. Choosing a suitable classification algorithm for building IDS is not an easy task. The best method is to test the performance of the different classification algorithms. This paper aims to present the result of evaluating different classification algorithms to build an IDS model in terms of confusion matrix, accuracy, recall, precision, f-score, specificity and sensitivity. Nevertheless, most researchers have focused on the confusion matrix and accuracy metric as measurements of classification performance. It also provides a detailed comparison with the dataset, data preprocessing, number of features selected, feature selection technique, classification algorithms, and evaluation performance of algorithms described in the intrusion detection system.

Many applications in the real world include optimizing specific targets, such as cost minimization, energy conservation, climate, and maximizing production, efficiency, and sustainability. The optimization problem is strongly non-linear with multifunctional landscapes under several dynamic, non-linear constraints in some instances. It is challenging to address those issues. Also, with the increasing strength of modern computers, simplistic brute force methods are still inefficient and unwanted. Practical algorithms are also vital for these implementations whenever possible. Cloud computing has become an essential and popular emerging computing environment that supports on-demand services and provides internet-based services. Cloud computing allows a range of services and tools to be easily accessed from anywhere in the world. Since cloud computing has global access to its services, there will always be threats and challenges facing its servers and services, such as; task scheduling, security, energy efficiency, network load, and other challenges. In the research area, many algorithms have been addressed to solve these problems. This paper investigates relevant analysis and surveys on the above topics, threats, and outlooks. This paper offers an overview of nature-inspired algorithms, their applications, and valuation, emphasizing cloud computing problems. Many problems in science and engineering can be viewed as optimization problems with complex non-linear constraints. Highly nonlinear solutions typically need advanced optimization algorithms, and conventional algorithms can have difficulty addressing these issues. Because of its simplicity and usefulness, nature-inspired algorithms are currently being used. There are nevertheless some significant concerns with computing and swarming intelligence influenced by evolution.

No abstract

The exponential growth of the Internet of Things (IoT) technology poses various challenges to the classic centralized cloud computing paradigm, including high latency, limited capacity, and network failure. Cloud computing and Fog computing carry the cloud closer to IoT computers in order to overcome these problems. Cloud and Fog provide IoT processing and storage of IoT items locally instead of sending them to the cloud. Cloud and Fog provide quicker reactions and better efficiency in conjunction with the cloud. Cloud and fog computing should also be viewed as the safest approach to ensure that IoT delivers reliable and stable resources to multiple IoT customers. This article discusses the latest in cloud and Fog computing and their convergence with IoT by stressing deployment's advantages and complexities. It also concentrates on cloud and Fog design and new IoT technologies, enhanced by utilizing the cloud and Fog model. Finally, transparent topics are addressed, along with potential testing recommendations for cloud storage and Fog computing, and IoT.