Cloud Network Research Articles

A Stacked Ensemble Approach to Generalize the Classifier Prediction for the Detection of DDoS Attack in Cloud Network

A Stacked Ensemble Approach to Generalize the Classifier Prediction for the Detection of DDoS Attack in Cloud Network

Databases on Modern Networks: A Decade of Research That Now Comes into Practice

Modern cloud networks are a fundamental pillar of data-intensive applications. They provide high-speed transaction (packet) rates and low overhead, enabling, for instance, truly scalable database designs. These networks, however, are fundamentally different from conventional ones. Arguably, the two key discerning technologies are RDMA and programmable network devices. Today, these technologies are not niche technologies anymore and are widely deployed across all major cloud vendors. The question is thus not if but how a new breed of data-intensive applications can benefit from modern networks, given the perceived difficulty in using and programming them. This tutorial addresses these challenges by exposing how the underlying principles changed as the network evolved and by presenting the new system design opportunities they opened. In the process, we also discuss several hard-earned lessons accumulated by making the transition first-hand.

Functional segments and software defined trends in enterprise networks

In the current scenarios most of the business growth comes from the user experience, while using the business applications. Any enterprise network performance depends on the fundamental process of securely serving the user with applications. The network usage patterns have completely changed from just data transfer, voice, and video transmissions to serving entire applications over the networks. With all the modern applications served over the private and public clouds the network resources utilization must be optimized and take full advantage of them which is not possible with current network architectures. The enterprises are currently shifting towards software defined implementation due to many reasons like consistent policy application across the network, run time analytics for troubleshooting the problems and making the network cloud computing ready. Software defined approaches provide centralized security monitoring and control, user data and identity-based segmentation over the network, dynamic recovery from network infrastructure failure and finally virtualization of almost everything. This paper is an overview of the functioning of enterprises in the networking perspective along with current software defined approaches being used in enterprise networking which will soon take over the traditional networks by solving the challenges present in the software defined networking.

Dynamic route optimization for IoT based intelligent waste collection vehicle routing system

Efficient waste management has been a major problem for municipalities in urban communities. The one important task is to find the optimum route for the waste collection vehicle with the optimum cost of transportation, reduced emissions, and improved city image. The proposed waste management system consists of three major parts, Smart Bins, Smart vehicles, and a cloud network for communication. The smart bin is embedded with sensors to measure the level of garbage in the bin. On overshooting the overflow threshold, the bins become active in the network and send signals to the server through the cloud network. The active bins are clustered based on their geographical locations. A dynamic optimum path is generated for each vehicle within each cluster. The research proposes a clustering-based efficient waste collection vehicle routing system. For Dynamic path optimization, a weighted multiple heuristic-based Optimum A*(Op-A*) algorithm is proposed which utilizes heuristics like distance, traffic, road quality, fuel status, the waste amount in the smart bin, and vehicle load capacity. This research focuses on implementing and testing the proposed Op-A* algorithm with a synthetic dataset. The results are compared with existing state-of-art algorithms. The improved result generated by the proposed algorithm demonstrates the significantly outperformed state-of-the-art benchmarks from the literature in terms of total transportation cost and path selection time for the waste collection vehicle routing system.

Design and development Virtual Doctor Robot for contactless monitoring of patients during COVID-19

The main objective of this paper is to design and develop a virtual doctor robot (VDR) that will operate on the command of the actual doctor available far away from the patient through new technology AI and IoT. It is not possible for doctors to be present everywhere and every time, especially in remote areas of India during COVID-19. As a result, in an emergency, many patients lose their lives because they couldn’t reach the hospital on time or the doctor couldn’t be available on time. In such a situation, virtual doctor robots play a vital role in healthcare with real-time data processing with machine learning algorithms. In this paper, technological innovation has created new opportunities to enhance doctor services without contact using robots that can assist patients in dealing with their illnesses and providing solutions. Virtual doctor robots can benefit healthcare by satisfying patients with more precise solutions. The virtual doctor robot design presented in this paper can rescue patients in remote areas during any pandemic like COVID-19. Through this virtual doctor robot, the doctors can observe and communicate with the patients through video calls and prescribe necessary medicines through scanner. The proposed VDR will help the doctors check the patient's pulse, body temperature, heart rate, etc. through different sensors mounted in the robot structure and send that data using the Wi-Fi network. The doctors will use an IoT-based panel to control and monitor the robot and patients. The control commands that are sent by the doctor to the VDR are sent online and the robot controller then receives these commands. The VDR also showcases other functions like alerting the battery status to remind one that the battery needs to be charged. It also stores the data related to diagnosed patients through cloud networking. The framework of VDR is based on an automated vehicle with a four-wheel drive for its movement and communication between the doctor and VDR is in real-time through the internet. This paper also discusses the future trends of Virtual Doctor robots for the health sector, humanoid robots for surgical help, and other activities of the healthcare sector.

Performance evaluation of multiple satellite rainfall data sets in central highlands of Abbay Basin, Ethiopia

ABSTRACT This study evaluates four satellite-based precipitation datasets with gauged rainfall observations at a daily and wet season time scales. Satellite Precipitation Estimators from Climate Hazards Group Infrared Precipitation Stations (CHIRPSv2), the Climate prediction center (CPC) morphing technique (CMORPH), the Integrated Multi-satellite Retrieval for GPM (IMERG-06) and Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks Cloud Classification System (PERSSIANN-CCS) were used. Categorical and continuous statistical techniques were applied to evaluate the detection and estimation ability of SPEs. Among the SPEs, CMORPH was superior, 0.90 and followed by IMERG-06, 0.82 in the Probability of detection (POD), Critical success index (CSI), 0.74 and 0.70 and Accuracy, 0.75 and 0.72 at wet season, respectively. The result of bias ratio showed underestimation for all SPEs at wet season and daily time scale. The comparison of SPEs in terms of RMSE indicated that IMERG-06 was better, 75.9 followed by CMORPH, 86.0. In terms of Relative bias (RBias), IMERG-06 was the second next to CHIRPSv2. On the other hand, IMERG-06 had better RMSE at JJAS and daily time scale. Overall, the findings of this evaluation study showed the capacity of IMERG-06 product could be reliable rainfall data sources for climate and hydrological analysis of central highlands of Abbay Basin.

Hilbert curves-based location privacy technique for vehicular cloud networks

Hilbert curves-based location privacy technique for vehicular cloud networks

An effective and efficient parallel large-scale cross-media retrieval in mobile cloud network

An effective and efficient parallel large-scale cross-media retrieval in mobile cloud network

IoT enabled vehicle recognition system using inkjet-printed windshield tag and 5G cloud network

An IoT-enabled vehicle recognition system is developed and tested in this work, besides the tag antenna design and prototype for the vehicle windshield. Firstly, a flexible, surface-tolerant, low-profile, and low-cost passive UHF-RFID windshield tag inlay is designed. The tag is prototyped as wet-inlay with silver conductive-ink inkjet printed antenna on Polyethylene-Terephthalate layer, bonded with Higgs-EC IC. The dimension of the antenna is 0.25 × 0.05λ0. The overall inlay structure is around 0.27 × 0.06λ0. The antenna is designed at ETSI band, with a return loss of −14 dB at 900 MHz and a read range of 13.3 m and 8 m over lower and upper bands, respectively. The antenna demonstrates an omnidirectional radiation pattern and a maximum gain of 1.92 dBi. The developed prototype is tested in both indoor and outdoor environments. Secondly, the data is sent over the IoT cloud where decision-making is made for authorization. The reader data is transmitted over MQTT broker, via the Scotland 5G network to the MongoDB database using Python module Pymongo. The algorithm for vehicle recognition is written and verified by testing the dummy registered tags for access authorization, with a 100% recognition accuracy. We achieved higher read ranges, surface tolerance, and low profile features, comparatively. The overall system shows optimally great results.

Analysis of Network Security in IoT-based Cloud Computing Using Machine Learning

Network security in IoT-based cloud computing can benefit greatly from the application of machine learning techniques. IoT devices introduce unique security challenges with their large-scale deployments and diverse nature. Machine learning can help address these challenges by analyzing IoT network traffic, detecting anomalies, identifying potential threats, and enhancing overall network security. The security of cloud networks is validated using binary classification to detect attacks. Random forest classifiers achieved an accuracy of 96%, while K nearest classifier had an accuracy of 93% and a precision value of 0.96. The proposed model ensures security of big data against intrusion attacks on the network. Although machine learning techniques can be powerful for protecting cloud computing networks, challenges still need to be addressed before widespread adoption. Understanding the potential and limitations of machine learning approaches to network security canhelp researchers and practitioners develop more effective strategies for safeguarding their systems in an increasingly interconnected world. Network security of big data in cloud computing can be enhanced by applying machine learning techniques. Machine learning algorithms can analyze large amounts of data to detect patterns, anomalies, and potential security threats. Here are several ways machine learning can be utilized to improve network security in the context of big data and cloud computing.

FlowPinpoint: Localizing Anomalies in Cloud-Client Services for Cloud Providers

For public cloud providers, it is of great significance to maintain the availability of their cloud services, which requires efficient anomaly diagnosis and recovery. To achieve such properties, the first step is to localize the anomalies, i.e., determining where they happen in the network path of cloud-client services. We propose FlowPinpoint to perform anomaly localization for cloud providers. FlowPinpoint collects statistics of each network flow at the cloud network gateways (i.e., gateway flowlog), where the collected data can reflect the information from both the cloud side and the Internet side. Aggregation and association are conducted on the datacenter-scale gateway flowlogs by Alibaba's big data computing platform. In order to preclude the disturbance of anomaly-unrelated flowlogs, a two-layer filter is proposed which consists of an indicator-based filter and an isolation forest filter. Finally, the anomaly localization analyzer classifies the flowlogs and determines whether the anomaly is inside the cloud network or not according to the classification results. FlowPinpoint is implemented and tested in the production environment of Alibaba Cloud, and it correctly localizes 1 anomaly inside the cloud and 6 anomalies on the Internet over 4 months.

Preface

The 2023 6th International Symposium on Big Data and Applied Statistics (ISBDAS 2023) was held in Shanghai, China from March 10th to 12th, 2023. The purpose of ISBDAS 2023 was to discuss the present status and future perspectives of big data and applied statistics technology. The following topics were selected in order to cover most of the scientific program and highlight an area where new ideas have emerged over recent years:(1) Big Data Mining and Visualization;(2) Big Data Security and Privacy;(3) Computational Mathematics;(4) Probability Theory;(5) Statistical Methodology, etc.There were about 60 participants. In addition to the keynote speeches, we had many talks selected from the contributed papers. There were invited, oral, and poster sessions. This proceedings contains the papers presented in invited and selected talks together with those presented in poster sessions.At the Conference, the participants discussed a wide range of issues affecting the theoretical and computational aspects of research problems in the big data and applied statistics technology. Besides, we had a comprehensive overview of this fascinating field and of future scenarios thanks to the participation of leaders of the most important projects. Professor Zhihui Lv from Fudan University, China performed a keynote speech and shared with us his own research experience and professional view. His main research fields include: network multimedia technology, cloud computing and service computing technology, big data architecture, edge computing, blockchain distributed system, network security technology, cloud data center management technology. And still another speaker was Jingsong Yang, a Professor at Shanghai Jiao Tong University, China. As a third-level researcher, he has undertaken more than 20 national, provincial and international cooperation projects, and has made many innovative achievements in the mechanism of marine microwave remote sensing, remote sensing technology and application, and remote sensing equipment development.We wish to acknowledge our host institutions, sponsors, and the many people whose work made this Conference possible. We would like to express our gratitude to the members of the International Technical Program Committee and those of the Organizing Committee for their efforts which made this Workshop successful. Particularly we would like to present our great thanks to the editors and other colleagues of Journal of Physics: Conference Series for their patience and kind assistance in publishing this paper volume.The Committee of ISBDAS 2023list of Committee Member is available in this pdf.

DDNSAS: Deep reinforcement learning based deep Q-learning network for smart agriculture system

As the global population continues to grow and environmental conditions become increasingly unpredictable, meeting the demands for food becomes increasingly difficult. To overcome these challenges, smart agriculture has emerged as the key technology. Deep Learning model with Internet of Things (IoT), unmanned aerial vehicle (UAV), and edge–fog–cloud architecture enabled smart agriculture as a key component for next agriculture revolution. In this work, we present two stage end-to-end DRL based smart agricultural system. In stage one, we proposed ACO enabled DQN (MACO-DQN) model to offload task including fire detection, pest detection, crop growth monitoring, irrigation scheduling, soil monitoring, climate monitoring, field monitoring etc. MACO-DQN model offload the task to either edge, fog or cloud networking devices based on latency, energy consumption and computing power. Once the task offloaded to computing devices (edge, fog or cloud), task of prediction and monitoring various agriculture activities is performed at stage two. In stage two, we proposed DRL based DQN (RL-DQN) model for prediction and monitoring agricultural task activities. Finally, we demonstrate experimental findings of our proposed model that represent a marked enhancement in terms of convergence speed, planning success rate, and path accuracy. To evaluate its performance, the method presented in this paper was compared to traditional deep Q-networks-based intensive learning method under consistent experimental conditions. Overall, 98.5% precision, 99.1% recall, 98.1% F-measure, and 98.5% accuracy is obtained when using our proposed methodology and based on the performance results the model outperforms other existing methodologies.

Discovering Key Actors and Opinion Leaders on Twitter’s Start-Up and Entrepreneurship Topics Trending: A Social Network Analysis Approach

The topic of start-ups and entrepreneurship has been widely discussed on social media to disseminate information while serving as a digital marketing platform. Unfortunately, researchers and practitioners interested in analyzing social networks to gain business advantage through digital platforms still need to be expanded. This paper analyzes networks on social media to find important actors and opinion leaders on trending start-up and entrepreneurship topics. Data analytics is performed using one of the most extensive social media, Twitter. Several visualizations of the social network of these actors are then displayed using specific network visualization applications. The results are discussed by analyzing the main attributes in the network, namely degree, centrality, modularity, and word cloud, which can lead us to discover important attributes in the network: main actors, opinion leaders, groupings of actors according to topics discussed, and specific terms that are trending topics. This paper contributes to digital marketing practitioners and social network analysis studies in related fields with a concise and duplicable methodology. Further research is recommended to work more extensively with more representative data from multiple networks to analyze these topics in more depth.

Distributed computing in special purpose wireless telecommunication systems: performance and prospects

The study is devoted to the analysis of the application of distributed computing technologies in wireless telecommunication systems. The paper considers the concept of cloud radio access networks (C-RAN) and their use to improve productivity and reduce costs of communication operators. The logical relationship of mobile devices, cloud wireless network and internal servers is considered, as well as the role of split-TCP-proxy in the distribution of connections and maintenance of constant communication between the user and the internal server of the cloud. It is noted that the dynamic operations of the wireless network cloud, such as topology configuration and data rate distribution, provide better service at the upper layer. The importance of knowing the state of the wireless channel for the successful implementation of operations in C-RAN is emphasized. It is noted that the use of distributed computing in wireless systems can reduce deployment and maintenance costs, providing efficient data transmission and improved performance. The obtained results indicate the possibility of successful use of distributed computing technologies in wireless telecommunication systems. This opens prospects for communication operators to reduce costs and improve the quality of service. The research is relevant in the context of the continuous development of wireless technologies and has the potential for further research and improvement of wireless telecommunication systems. Future work may focus on expanding the scope of research, including a wider range of technologies and algorithms that can be used to optimize the performance of cloud radio access networks. In addition, the impact of using distributed computing technologies on other aspects of wireless systems, such as security, resource management, and scalability, can be considered. Research may also focus on the use of other protocols and algorithms to distribute computing tasks and optimize data transmission.

Analysis of Intrusion Detection Systems in UNSW-NB15 and NSL-KDD Datasets with Machine Learning Algorithms

Recently, the need for Network-based systems and smart devices has been increasing rapidly. The use of smart devices in almost every field, the provision of services by private and public institutions over network servers, cloud technologies and database systems are almost completely remotely controlled. Due to these increasing requirements for network systems, malicious software and users, unfortunately, are increasing their interest in these areas. Some organizations are exposed to almost hundreds or even thousands of network attacks daily. Therefore, it is not enough to solve the attacks with a virus program or a firewall. Detection and correct analysis of network attacks is vital for the operation of the entire system. With deep learning and machine learning, attack detection and classification can be done successfully. In this study, a comprehensive attack detection process was performed on UNSW-NB15 and NSL-KDD datasets with existing machine learning algorithms. In the UNSW-NB115 dataset, 98.6% and 98.3% accuracy were obtained for two-class and multi-class, respectively, and 97.8% and 93.4% accuracy in the NSL-KDD dataset. The results prove that machine learning algorithms are lateral to the solution in intrusion detection systems.

Two Heads Are Better than One: Image-Point Cloud Network for Depth-Based 3D Hand Pose Estimation

Depth images and point clouds are the two most commonly used data representations for depth-based 3D hand pose estimation. Benefiting from the structuring of image data and the inherent inductive biases of the 2D Convolutional Neural Network (CNN), image-based methods are highly efficient and effective. However, treating the depth data as a 2D image inevitably ignores the 3D nature of depth data. Point cloud-based methods can better mine the 3D geometric structure of depth data. However, these methods suffer from the disorder and non-structure of point cloud data, which is computationally inefficient. In this paper, we propose an Image-Point cloud Network (IPNet) for accurate and robust 3D hand pose estimation. IPNet utilizes 2D CNN to extract visual representations in 2D image space and performs iterative correction in 3D point cloud space to exploit the 3D geometry information of depth data. In particular, we propose a sparse anchor-based "aggregation-interaction-propagation'' paradigm to enhance point cloud features and refine the hand pose, which reduces irregular data access. Furthermore, we introduce a 3D hand model to the iterative correction process, which significantly improves the robustness of IPNet to occlusion and depth holes. Experiments show that IPNet outperforms state-of-the-art methods on three challenging hand datasets.

Enhanced resource allocation in distributed cloud using fuzzy meta-heuristics optimization

Cloud computing is a modern technology that has become popular today. A large number of requests has made it essential to propose a resources allocation framework for arriving requests. The network can be made more efficient and less costly this way. The cloud–edge paradigm has been considered a growing research area in the computing industry in recent years. The increase in the number of customers and requests for cloud data centers (CDCs) has created the need for robust servers and low power consumption mechanisms. Ways to reduce energy in the CDC having appropriate algorithms for resource allocation. The purpose of this study was to develop an intelligent method for dynamic resource allocation using Takagi–Sugeno–Kang (TSK) neural-fuzzy systems and ant colony optimization (ACO) techniques to reduce energy consumption by optimizing resource allocation in cloud networks. It predicts future loads using a drop-down window to track CPU usage. By optimizing virtual machine migration, ACO can reduce energy consumption. Simulations are provided by examining the implementation and a variety of parameters such as the number of requests made wasted resources, and requests rejected. In this paper, we propose the use of virtual machine migration to accomplish two main goals: evacuating additional and non-optimal virtual machines (scaling and shutting down additional active physical machines) and solving the resource granulation problem. We evaluated and compared our results with literature for rejection rates of virtual and physical machine applications. The performances of our algorithms are compared to different criteria such as performance in request rejection, dynamic CPU resource allocation with reinforcement learning, multi-objective resource allocation, NSGAIII, Whale optimization and Forecast Particle Swarm allocation. A comparison of some evaluation criteria showed that the proposed method is more efficient than other methods.

Reveal: Robustness-aware VNF placement and request scheduling in edge clouds

In the edge cloud network, service providers place virtual network functions (VNFs) in edge clouds to serve users’ requests. Thus, it is essential to consider VNF placement and request scheduling in edge clouds. Existing works often focus on minimizing request completion time or maximizing network throughput to utilize network resources and ensure users’ QoS efficiently. However, they ignore two practical factors: malicious users and failed VNFs, leading to poor network robustness. To this end, this paper studies robustness-aware VNF placement and request scheduling, named Reveal. Specifically, we limit the number of VNFs each user can access and the number of users each VNF can serve to control the influence scope of malicious users and VNF failures. Since placing VNFs is time-consuming and requests arrive dynamically, we solve this problem through two phases: robust VNF placement and assignment, and online request scheduling. For the first phase, we design an efficient knapsack-based rounding algorithm with bounded approximation factors. For online request scheduling, we propose a primal–dual based algorithm with a competitive ratio of 1−ϵ,O(log1/ϵ) where ϵ∈(0,1). Experiment and simulation results show that Reveal can achieve better performance and robustness than other alternatives.

Rancang Bangun Peringatan Dini Bahaya Banjir Pada Daerah Aliran Sungai Berbasis Internet Of Think

High rainfall, High rainfall, especially during the rainy season, results in a rise in the level of water flowing into the river, which can be a triggering factor for the danger of flooding. One indication that can be known is the overflow of water in the river flow so that the flowing water cannot be accommodated by the river canal. In this study, a monitoring system can provide water level reading information using the HC-SR04 ultrasonic sensor with the addition of a flow meter sensor reading feature, and a DHT11 temperature sensor. This flood early warning system is designed to provide a warning when a flood hazard will occur which is indicated by an indication of rising water levels in the river flow with information directly or in real time as a disaster early warning effort. This information will then be displayed on a 16x2 LCD, alarm, and smartphone direct display via Internet of Think technology with ESP8266 as the transmission medium to the cloud network. In addition to presenting information, the tool is also equipped with a servo motor that will activate the opening of the sluice gate so that the water level returns to normal. In the appearance of this information later, the system will display the state of the level of early warning information with a red LED indicator display and a buzzer that lights up when conditions are dangerous and can be accessed via a smartphone. With this tool designed later, it is hoped that it will be able to anticipate the danger of flooding that will occur so that it can help residents around the riverbanks to get early warning when a possible flood hazard will occur.