RETRACTED: Detection of anomalies in cloud services using network flowdata analysis

Cloud computing is a new computing model which uses virtualization technology, distributed computing, parallel computing and other existing technologies to achieve cloud service virtualization and economies of scale, whilst increasingly overwhelming cloud security issues has brought great challenges and concerns to the cloud services providers and cloud users, especially trust and privacy issues with regard to cloud computing and cloud shared storage associated security issues. In the paper, we expound the basic concepts of cloud computing, deployment models, service models and key features, analyze and outline the currently highlighted cloud security issues, report the status quo of cloud computing security, investigate the prevalent and typical cloud computing security problem key solving techniques, and thus render a comprehensive cloud computing security technical reference model, which is composed of associated cloud security solving techniques that result from inevitably multi-faceted cloud security issues. The model is expected to alleviate prominent cloud security issues. This paper generalizes cloud security technology research directions and further development space of cloud security technology and standardization.KeywordsCloud computingSecurity issuesTrusted cloud computing solving techniquesShared storage security solving techniquesCloud security technical reference model

Cloud computing and big data: Technologies and applications

Privacy-preserving multi-party logistic regression in cloud computing

The use of Cloud Computing can increase the service efficiency and Service Level Agreement for cloud users by linking them to the appropriate cloud service provider using cloud services brokerage paradigm. Cloud services brokerage represents a new promising layer to be added to cloud computing network, which manages the use, performance and delivery of cloud services, and negotiates relationships between cloud services providers and cloud services consumers. The work presented in this paper studies the research related to cloud service brokerage systems along with the weaknesses and drawbacks associated with each of these systems, with a particular focus on the multi-cloud-based services environment. In addition, the paper will end up with a proposed multi-cloud framework that overcomes the weaknesses of the other listed cloud brokers. The new framework aims at finding the appropriate data centre in terms of energy efficiency, QoS, and SLA. Moreover, it highlights the entire key features that must be available in the multi-cloud-based brokerage systems.

Internet of things intrusion detection model and algorithm based on cloud computing and multi-feature extraction extreme learning machine

Internet of Things (IoT) concept provides a number of opportunities to improve our daily lives while also creating a potential risk of increasing the vulnerability of personal information to security and privacy breaches. Data collected from IoT is usually offloaded to the Cloud which may further leave data prone to a variety of attacks if security and privacy issues are not handled properly. Anomaly detection has been one of the widely adopted security measures in wired and wireless networks. However, it is not straight forward to apply most of the anomaly detection techniques to IoT and cloud. One of the main challenges is deriving outlier features from the vast volume of data pumped from IoT to the cloud. Other challenges include the large number of sources generating data, heterogenous connectivity and traffic patterns of IoT devices, cloud services being offered at geographically remote places and causing IoT data to be stored in different countries with different legislations. This paper, for the first time, presents the challenges and opportunities in anomaly detection for IoT and cloud. It first introduces the prominent features and application fields of IoT and Cloud, then discusses security and privacy risks to personal information and finally focuses on solutions from anomaly detection perspective.

The amount of Internet data is significantly increasing due to the development of network technology, inducing the appearance of big data. Experiments have shown that deep mining and analysis on large datasets would introduce great benefits. Although cloud computing supports data analysis in an outsourced and cost-effective way, it brings serious privacy issues when sending the original data to cloud servers. Meanwhile, the returned analysis result suffers from malicious inference attacks and also discloses user privacy. In this paper, to conquer the above privacy issues, we propose a general framework for Preserving Multiparty Data Privacy (PMDP for short) in cloud computing. The PMDP framework can protect numeric data computing and publishing with the assistance of untrusted cloud servers and achieve delegation of storage simultaneously. Our framework is built upon several cryptography primitives (e.g., secure multiparty computation) and differential privacy mechanism, which guarantees its security against semihonest participants without collusion. We further instantiate PMDP with specific algorithms and demonstrate its security, efficiency, and advantages by presenting security analysis and performance discussion. Moreover, we propose a security enhanced framework sPMDP to resist malicious inside participants and outside adversaries. We illustrate that both PMDP and sPMDP are reliable and scale well and thus are desirable for practical applications.

Cloud computing, and cloud services in particular, offer the Research and Education sector huge opportunities to both maximise effectiveness and reduce the capital investment and development time to deliver results. By utilising shared and off-the-shelf services for commodity activities, the R&E community can refocus its design, development and support resources into those fields that cannot be easily provided by the commercial sector. Cloud computing empowers users to select and use the services they really want, in an easy and often economically attractive manner. The broad standardisation of service delivery offers substantial advantages with scalability and user acceptance. By using services that the users have had experience of outside the R&E community, training requirements can be minimised and personal efficiency can be improved. The scalability of cloud services also allows rapid expansion or contraction of capacity as the project requires with minimal penalties. This near-linear cost model allows easier budgeting and financial control.

Over the past few years, cloud computing has been more and more attractive as a new computing paradigm due to high flexibility for provisioning on-demand computing resources that are used as services through the Internet. In cloud computing, the unique characteristics in cloud services such as dynamic and diverse services offered at different levels, together with the lack of standardized description languages pose significant challenges for effective cloud service discovery. In this paper, we propose a cloud service search engine that exploits a novel ontology-based technique for identifying cloud service categories to improve the accuracy of cloud services searching in real environments. Our approach has the capability to automatically identify and categorize cloud services by detecting cloud service concepts from cloud service sources. Initially, we focus on building the cloud service ontology by using NIST (US National Institute of Standards and Technology). Then, we utilize our cloud service categorization method to investigate cloud service ontology's concepts in a real-world cloud services dataset which contains the metadata of 5,883 real cloud services. After that, we generate cloud service clusters by using cosine similarity to build the cloud service categorization. Our cloud service categorization is helpful in determining whether a given web source is a cloud service. Furthermore, the new web resource which has been categorized as a cloud service can be used to boost knowledge of cloud service categorization cumulatively. The proposed approach is validated by using real-world cloud services available on the World Wide Web and the experimental results show the effectiveness of the approach in cloud service discovery.

Destructive cyber-attacks and cybercriminals are increasing with the increase in IoT (Internet of Things) devices globally. This ha/s led to the need for increase in security in IoT systems. Innovative and novel techniques are used by the intruders to accomplish malicious goals effectively through cyber-attacks. An Intrusion Detection System (IDS) is used for classification of attacks in IoT networks based on anomaly detection and machine learning techniques. Inefficiency is observed in the conventional machine learning models and intrusion detection techniques as the network technologies are unpredictable. Accurate identification of various anomalies is possible with deep learning models in several research segments. The input data along with its prominent characteristics may be categorized automatically for classification and anomaly detection using convolutional neural networks (CNN). Faster computations are enabled due to the performance efficiency of CNN. For IoT networks, an intrusion detection model based on anomaly detection is designed and developed in this paper. A multiclass classification framework is created initially using a CNN model. Further, 3D CNN is used for implementation of the proposed model. Various intrusion detection datasets from IoT networks are used for validation of the proposed CNN model. Pre-trained multiclass CNN model is used for implementation of multiclass and binary classification based on transfer learning. When compared to the conventional deep learning models, the proposed multiclass and binary classification framework has attained improved F1 score, recall, precision and accuracy.

Cloud computing faces numerous challenges in many areas including security and privacy issues. In this work, a developed approach is suggested to tackle three security and privacy issues: network intrusion detection (NID), privacy, and internal attacks. A decision tree (J48) has been used to generate a set of rules based on the CICIDS2017 dataset to solve the NID problem. The accuracy of the generated rules approaches 99.8%. A set of policies are attached to the data file on the bases of a sticky policy to preserve privacy. A new approach is suggested based on blockchain to detect internal attacks in real-time, in which a set of trustees-chain are identified by the data owner. Any data modification conducted by a trusted member will be reported to all members of the trust group including the owner. The developed approach suggests adding a Privacy and Detecting Intrusions Service (PDIS) layer as part of the cloud computing main service model. PDIS includes the three suggested approaches above (NID, sticky policy, and trustees-chain). Finally, a web-based application is implemented to act as casework to validate PDIS and evaluate its reliability. Keywords: Cloud computing, Privacy, Machine learning, Internal Intrusion Detection, Network Intrusion Detection, Real Time Auditing

Cloud computing is a framework for delivering services through a shared network. It should follow certain procedures or mechanisms to protect and make the cloud data more reliable. In secure cloud computing, the implementation and service delivery model are dealt with. Cloud security could be viewed as a provider and customer responsibility depending on their design and level. There are cases where multiple cloud users can use the same information to create service abnormalities, which is called ‘multitenancy.' In a multi-tenant model, the resources are isolated between different users and it is done by segmentation to achieve the goal of data security between many users. In the user access area, user access to common resources is also limited. The software is aware of the actions of the client and acts accordingly if unauthorized access is detected. The authors propose a clear idea of the data storage security of cloud information through honeypot in cloud monitoring applications and various cloud security scheme key models used and also identifies the audit time and invalid response time.

An increasing number of applications are being hosted on cloud based platforms. Cloud platforms are serving as a general computing facility and applications being hosted on these platforms range from simple multi-tier web applications to complex social networking, eCommerce and Big Data applications. High availability, performance and auto-scaling are key requirements of Cloud based applications. Cloud platforms serve these requirements using dynamic provisioning of resources in on-demand, multi-tenant fashion. A key challenge for cloud service providers is to ensure the Quality of Service (QoS), as a user / customer requires more explicit guarantees of QoS for provisioning of services. Cloud service performance problems can directly lead to extensive financial loses. Thus, control and verification of QoS become a vital concern for any production level deployment. Therefore, it is crucial to address performance as a managed objective. The success of cloud services depends critically on automated problem diagnostics and predictive analytics enabling organizations to manage their performance proactively. Moreover, effective and advance monitoring is equally important for performance management support in clouds. In this thesis, we explore the key techniques for developing monitoring and performance management systems to achieve robust cloud systems. At first, two case studies are presented as a motivation for the need of a scalable monitoring and analytics framework. It includes a case study on performance issues of a software service, which is hosted on a virtualized platform. In the second case study, cloud services are analyzed that are offered by a large IT service provider. A generalization of case studies forms the basis for the requirement specifications which are used for state-of-the-art analysis. Although, some solutions for particular challenges have already been provided, a scalable approach for performance problem diagnosis and prediction is still missing. For addressing this issue, a distributed scalable monitoring and analytics framework is presented in the first part of this thesis. We conducted a thorough analysis of technologies to be used by our framework. The framework makes use of existing monitoring and analytics technologies. However, we develop custom collectors to retrieve data non-intrusively from different layers of cloud. In addition, we develop the analytics subscriber and publisher components to retrieve service related events from different APIs and sends alerts to the SLA Management component for taking corrective measures. Further, we implemented an Open Cloud Computing Interface (OCCI) monitoring extension using OCCI Mixin mechanism. To deal with performance problem diagnosis, a novel distributed parallel approach for performance anomaly detection is presented. First all anomalous metrics are found from a distributed database of time-series for a particular window. For comparative analysis three light-weight statistical anomaly detection techniques are selected. We extend these techniques to work with MapReduce paradigm and assess and compare the methods in terms of precision, recall, execution time, speedup and scale up. Next, we correlate the anomalous metrics with the target SLO in order to locate the suspicious metrics. We implemented and evaluated our approach on a production Cloud encompassing Infrastructure as a Service (IaaS) and Platform as a Service (PaaS) service models. Experimental results confirm that our approach is efficient and effective in capturing the metrics causing performance anomalies. Finally, we present the design and implementation of an online anomaly prediction system for cloud computing infrastructures. We further present an experimental evaluation of a set of anomaly prediction methods that aim at predicting upcoming periods of high utilization or poor performance with enough lead time to enable the appropriate scheduling, scaling, and migration of virtual resources. Using real data sets gathered from Cloud platforms of a university data center, we compare several approaches ranging from time-series (e.g. auto regression (AR)) to statistical classification methods (e.g. Bayesian classifier). We observe that linear time-series models, especially AR models, are most likely suitable to model QoS measures and forecast their future values. Moreover, linear time-series models can be integrated with Machine Learning (ML) methods to improve proactive QoS management.

Ensuring the security of cloud computing is one of the most critical challenges facing the cloud services sector. Dealing with data in a cloud environment, which uses shared resources, and providing reliable and secure cloud services, requires a robust encryption solution with no or low negative impact on performance. Thus, this study proposes an effective cryptography solution to improve security in cloud computing with a very low impact on performance. A complex cryptography algorithm is not helpful in cloud computing as computing speed is essential in this environment. Therefore, this solution uses an improved Blowfish algorithm in combination with an elliptic-curve-based algorithm. Blowfish will encrypt the data, and the elliptic curve algorithm will encrypt its key, which will increase security and performance. Moreover, a digital signature technique is used to ensure data integrity. The solution is evaluated, and the results show improvements in throughput, execution time, and memory consumption parameters.

Privacy preserving multi-party computation delegation for deep learning in cloud computing