Zhigang Hu

Toward trustworthy cloud service selection

Cloud services consumers face a critical challenge in selecting trustworthy services from abundant candidates, and facilitating these choices has become a critical issue in the uncertain cloud industry. This paper employs the time series analysis to address challenges resulting from fluctuating quality of service, flexible service pricing and complicated potential risks in order to propose a time-aware trustworthy service selection approach with tradeoffs between performance-costs and potential risks. The original evaluation data about the services is preprocessed using a cloud model, and interval neutrosophic set (INS) theory is utilized to describe and measure the performance-costs and potential risks of services. In order to calculate and compare the candidate services while supporting tradeoffs between performance-costs and potential risks in different time periods, we established a cloud service interval neutrosophic set (CINS) and designed its operators and calculation rules, with theoretical proofs provided. The problem of time-aware trustworthy service selection is formulated as a multi-criterion decision-making (MCDM) problem of creating a ranked services list using CINS, and it is solved by developing a CINS ranking method. Finally, experiments based on a real-world dataset illustrate the practicality and effectiveness of the proposed approach. Propose a time-aware service selection approach for uncertain cloud industry.Formulate a multi-criterion decision-making problem using interval neutrosophic set.Support tradeoffs between performance-costs and potential risks in time periods.Establish the CINS theory to calculate and compare the candidate cloud services.Develop a CINS ranking method to create a ranked list of trustworthy cloud services.

Virtual Machine Placement Algorithm for Both Energy-Awareness and SLA Violation Reduction in Cloud Data Centers

The problem of high energy consumption is becoming more and more serious due to the construction of large-scale cloud data centers. In order to reduce the energy consumption and SLA violation, a new virtual machine VM placement algorithm named ATEA adaptive three-threshold energy-aware algorithm, which takes good use of the historical data from resource usage by VMs, is presented. In ATEA, according to the load handled, data center hosts are divided into four classes: hosts with little load, hosts with light load, hosts with moderate load, and hosts with heavy load. ATEA migrates VMs on heavily loaded or little-loaded hosts to lightly loaded hosts, while the VMs on lightly loaded and moderately loaded hosts remain unchanged. Then, on the basis of ATEA, two kinds of adaptive three-threshold algorithm and three kinds of VMs selection policies are proposed. Finally, we verify the effectiveness of the proposed algorithms by CloudSim toolkit utilizing real-world workload. The experimental results show that the proposed algorithms efficiently reduce energy consumption and SLA violation.

Time-aware trustworthiness ranking prediction for cloud services using interval neutrosophic set and ELECTRE

Abstract The imprecise quality of service (QoS) evaluations from consumers may lead to the inappropriate prediction for the trustworthiness of cloud services in an uncertain cloud environment. The service ranking prediction is a promising idea to overcome this deficiency of values prediction approaches by probing the ordering relations between cloud services concealed in the imprecise evaluations. To address the challenges for trustworthy service selection resulting from fluctuating QoS, flexible service pricing and complicated potential risks, this paper proposes a time-aware approach to predict the trustworthiness ranking of cloud services, with the tradeoffs between performance-cost and potential risks in multiple periods. In this approach, the interval neutrosophic set (INS) theory is utilized to describe and assess the performance-costs and potential risks of cloud services: (1) the original evaluation data about cloud services are preprocessed into the trustworthiness interval neutrosophic numbers (INNs); (2) the new INS operators are proposed with the theoretical proofs to calculate the possibility degree and the ranking values of trustworthiness INNs, contributing to the identification of the neighboring users based on the Kendall rank correlation coefficient (KRCC). The problem of time-aware trustworthiness ranking prediction is formulated as a multi-criterion decision-making (MCDM) problem of creating a ranked services list using INS, and an improved ELECTRE method is developed to solve it. The proposed approach is verified by experiments based on an appropriate baseline for comparative analysis. The experimental results demonstrate that the proposed approach can achieve a better prediction quality than the existing approach. The results also show that our approach works effectively in the risk-sensitive and performance-cost-sensitive application scenarios and prevent the malignant price competition launched by low-quality services.

User feature-aware trustworthiness measurement of cloud services via evidence synthesis for potential users

Cloud computing can provide elastic and dynamic resources on demand, which facilitates service providers to make profits resulting from the long tail effect. It becomes vitally important to ensure that cloud services can be acceptable to more potential users. However, it is challenging for potential users to discover the trustworthy cloud services due to the deficiency of usage experiences and the information overload of QoE (quality of experience) evaluations from consumers. This paper presents a user feature-aware trustworthiness measurement approach for potential users. In this approach, the influence factors of QoE are systematically analyzed based on the user feature model and the quantitative computation methods are designed to measure the user feature similarity. In addition, employing FAHP (fuzzy analytic hierarchy process) method identifies the user feature community. To enhance the accuracy of trustworthiness measurement, the false evidences in QoE evaluations are iteratively filtered out with dynamic mean distance threshold. Finally, the trustworthiness of service is measured via evidence synthesis combining user feature similarity. The experiments show that this approach is effective to improve the quality of trustworthiness measurement, which is helpful to solve information overload problem and cold start problem of trusted service recommendation for potential users. źThe similarity measurement methods for six user features are proposed respectively.źThe weights of user features are assigned appropriately by employing the FAHP method.źThe service trustworthiness is measured via evidence theory with feature weights.

Minimizing SLA violation and power consumption in Cloud data centers using adaptive energy-aware algorithms

Abstract In this paper, we address the problem of reducing Cloud datacenter high energy consumption with minimal Service Level Agreement (SLA) violation. Although there are many energy-aware resource management solutions for Cloud datacenters, existing approaches focus on minimizing energy consumption while ignoring the SLA violation at the time of virtual machine (VM) deployment. Also, they do not consider the types of application running in the VMs and thus may not really reduce energy consumption with minimal SLA violation under a variety of workloads. In this paper, we propose two novel adaptive energy-aware algorithms for maximizing energy efficiency and minimizing SLA violation rate in Cloud datacenters. Unlike the existing approaches, the proposed energy-aware algorithms take into account the application types as well as the CPU and memory resources during the deployment of VMs. To study the efficacy of the proposed approaches, we performed extensive experimental analysis using real-world workload, which comes from more than a thousand PlanetLab VMs. The experimental results show that, compared with the existing energy-saving techniques, the proposed approaches can effectively decrease the energy consumption in Cloud datacenters while maintaining low SLA violation.

Multi-valued collaborative QoS prediction for cloud service via time series analysis

Abstract Aiming at the diversity of user features, the uncertainty and the variation characteristics of quality of service (QoS), by exploiting the continuous monitoring data of cloud services, this paper proposes a multi-valued collaborative approach to predict the unknown QoS values via time series analysis for potential users. In this approach, the multi-valued QoS evaluations consisting of single-value data and time series data from consumers are transformed into cloud models, and the differences between potential users and other consumers in every period are measured based on these cloud models. Against the deficiency of existing methods of similarity measurement between cloud models, this paper presents a new vector comparison method combining the orientation similarity and dimension similarity to improve the precision of similarity calculation. The fuzzy analytic hierarchy process method is used to help potential users determine the objective weight of every period, and the neighboring users are selected for the potential user according to their comprehensive similarities of QoS evaluations in multiple periods. By incorporating the multi-valued QoS evaluations with the objective weights among multiple periods, the predicted results can remain consistent with the periodic variations of QoS. Finally, the experiments based on a real-world dataset demonstrate that this approach can provide high accuracy of collaborative QoS prediction for multi-valued evaluations in the cloud computing paradigm.

A Hybrid Skin Detection Model from Multiple Color Spaces Based on a Dual-Threshold Bayesian Algorithm

As a preliminary step of many applications, skin detection serves as an irreplaceable role in image processing applications, such as face recognition, gesture recognition, web image filtering, and image retrieval systems. Combining information from multiple color spaces improves the recognition rate and reduces the error rate because the same color is represented differently in other color spaces. Consequently, a hybrid skin detection model from multiple color spaces based on a dual-threshold Bayesian algorithm (DTBA) has been proposed. In each color space, the pixels of images are divided into three categories, namely, skin, nonskin, and undetermined, when using the DTBA. Then, nearly all skin pixels are obtained by using a specific rule that combines the recognition results from multiple color spaces. Furthermore, skin texture filtering and morphological filtering are applied to the results by effectively reducing false identified pixels. In addition, the proposed skin model can overcome interference fr...

An energy conservation replica placement strategy for Dynamo

With the fast development of cloud computing and wide application of cloud storage, the energy efficiency of cloud storage system is drawing significant attention from researchers or specialists. For the typical Dynamo cloud storage system, we design a new policy instead of the consistent hashing policy which is a combination of consistent hashing and sequential policy. The basic idea of this policy is that it divides the nodes into groups and allows each other to be mirror modes so it can find the full coverage subset of data items easily. Also we use autoregressive-moving-average model to estimate the task numbers of servers so that when under low utilization period, certain numbers of servers can be turned off to save energy. Based on the model, we demonstrate that it can save up to 23.7xa0% energy and maintain load balancing of servers. Furthermore, we compare our policy with Heuristic which is a classical energy conservation policy for cloud storage system that is based on consistent hash table. And we find several advantages of our policy which include finding the minimum subset of full coverage as well as other aspects.

A modified PSO algorithm for task scheduling optimization in cloud computing

With the increasing scale of tasks in cloud computing, the problem of high energy consumption becomes increasingly serious. To deal with the problem, we propose a cloud computing energy consumption model, which takes into account the execution and transmission cost of the processor. Then, based on this model, we put forward a task scheduling optimization algorithm named modified particle swarm optimization (M‐PSO) to handle the local optimum and slow convergence problem. Different from the PSO, M‐PSO can dynamically adjust the inertia weight coefficient to improve the speed of convergence according to the number of iterations. Finally, the performance of the proposed algorithm is evaluated through the CloudSim toolkit, and the experimental results show that the M‐PSO can efficiently reduce total cost compared with other algorithms.

Fine-Grained Energy Consumption Model of Servers Based on Task Characteristics in Cloud Data Center

In this paper, we address the problem of accurately modeling the cloud data center energy consumption. As minimizing energy consumption has become a crucial issue for the efficient operation and management of cloud data centers, an energy consumption model plays an important role in cloud datacenter energy management and control. Moreover, such model is essential for guiding energy-aware algorithms, such as resource provisioning policies and virtual machine migration policies. To this end, we propose a holistic cloud data center energy consumption model that is based on the principal component analysis and regression methods. Unlike the exiting approaches that focus on single system component in the datacenter, the proposed approach takes into account the energy consumption of the processing unit, memory, disk, and network interface card as well as the application characteristics. The proposed approach is validated through extensive experiments with the SPECpower benchmark. The experimental results show that the proposed energy consumption model achieves more than 95% prediction accuracy.