International Journal of Web and Grid Services (8 papers in press)
Managing Workflows on top of a Cloud Computing Orchestrator for using heterogeneous environments on e-Science
by Abel Carrión, Miguel Caballer, Ignacio Blanquer, Nelson Kotowski
Abstract: Scientific Workflows (SWFs) are widely used to model processes in e-Science. SWFs are executed by means of Workflow Management Systems (WMSs), which orchestrate the workload on top of computing infrastructures. The advent of cloud computing infrastructures has opened the door of using on-demand infrastructures to complement or even replace local infrastructures. However, new issues have arisen, such as the integration of hybrid resources or the compromise between infrastructure reutilization and elasticity. In this article we present an ad-hoc solution for managing workflows exploiting the capabilities of cloud orchestrators to deploy resources on demand according to the workload and to combine heterogeneous cloud providers (such as on-premise clouds and public clouds) and traditional infrastructures (clusters) to minimize costs and response time. The work does not propose yet another WMS, but demonstrates the benefits of the integration of cloud orchestration when running complex workflows. The article shows several configuration experiments from a realistic comparative genomics workflow called Orthosearch, to migrate memory-intensive workload to public infrastructures while keeping other blocks of the experiment running locally. The article computes running time and cost suggesting best practices.
Keywords: Workflow; Workflow Management Systems; Cloud Orchestrator; Multi-platform; e-Science; Cloud Computing; Comparative genomics.
Clustering-based uncertain QoS prediction of Web services via collaborative filtering
by Guobing Zou, Zhimin Zhou, Sen Niu, Yanglan Gan, Bofeng Zhang
Abstract: Although collaborative filtering (CF) has been widely applied for QoS-aware Web service recommendation, most of these approaches mainly focus on certain QoS prediction of Web services. However, they failed to take the natural characteristic of Web services with QoS uncertainty into account in real-world service-oriented Web applications. To solve the problem, this paper proposes a novel approach for uncertain QoS prediction via collaborative filtering and service clustering. We first establish uncertain QoS model for a service user by a tree-layer tree, where each service is formalized as a QoS matrix. To mine the similar neighborhood users for an active user, we then extend Euclidean distance to calculate the similarity between two uncertain QoS models. Finally, we present two novel QoS prediction strategies of QoS prediction based on collaborative filtering and service clustering, called U-Rec and UC-Rec. Extensive experiments have been conducted on large-scale real-world dataset that has more than 1.5 million uncertain QoS transaction logs of Web services. The experimental results demonstrate the effectiveness of our proposed approach.
Keywords: Web service; uncertain QoS prediction; collaborative filtering; service clustering.
Skyline Service Selection Approach based on QoS Prediction
by Yan Guo, Shangguang Wang, Kok-Seng Wong, Myung Ho Kim
Abstract: The Internet currently hosts a large number of Web services with highly volatile quality of service (QoS), which makes it difficult for users to quickly access highly reliable online services. Hence, the selection of the optimal service composition based on fast and reliable QoS has emerged as a challenging and popular problem in the field of service computing. In this paper, we propose a service selection approach based on QoS prediction. We consider historical QoS information as time series and predict QoS values using the autoregressive integrated moving average model, which can provide more accurate QoS attribute values. We then calculate the uncertainty in the prediction results using an improved coefficient of variation to prune redundant services. In order to downsize the search space, we employ Skyline computing to prune redundant services and perform Skyline service selection by using 0-1 mixed-integer programming. Experimental results based on real-world dataset showed that our approach yields satisfactory performance in terms of reliability and efficiency.
Keywords: service selection; QoS prediction; autoregressive integrated moving average model; Skyline service.
An Overall Approach to Achieving Load-Balancing for Hadoop Distributed File System
by Chi-Yi Lin, Ying-Chen Lin
Abstract: Hadoop Distributed File System (HDFS) is a popular cloud storage system that can scale up easily to meet the increasing demand for more storage capacity. In HDFS, files are divided into fixed-size blocks, which are then replicated and randomly stored on many DataNodes to prevent data loss. It can be easily observed that the random nature of the default block placement strategy may lead to a load imbalance state among the DataNodes. Although HDFS has a built-in utility to achieve load balancing, it comes at the cost of a reduced system performance owing to moving blocks around. In this paper, we take a holistic approach to achieving load balancing by considering all situations that may influence the load-balancing state. We designed a new role named BalanceNode to help in matching heavy-loaded and light-loaded DataNodes, so those light-loaded nodes can share part of the load from heavy-loaded ones. We also designed a better block placement strategy to make the storage load as balanced as possible in the first place. The simulation results show that our approach can achieve better load-balancing state than with existing algorithms.
Keywords: cloud computing; Hadoop Distributed File System; load balancing.
Real-time Adaptive QoS Prediction Using Approximate Matrix Multiplication
by Marin Silic, Adrian Kurdija, Sinisa Srbljic
Abstract: We introduce a novel QoS prediction model as a real-time support for selection of atomic service candidates based on their QoS properties while constructing composite applications. The proposed approach satisfies the following requirements: (i) fast and accurate prediction of QoS values, and (ii) adaptability with respect to environment changes. The model precomputes the similarities between users and services using approximate matrix multiplication to reduce the time complexity. When calculating a prediction for a user-service pair, the model considers similar users and services, but enhances the prediction accuracy by incorporating the number of observed records. Time complexity is further reduced by storing the lists of similar users and services which are updated in real-time. The model adapts to the changing environment: newer records are set to have greater influence on the predictions. The experiments conducted on relevant service-oriented datasets show advantages of the proposed model in accuracy and time performance.
Keywords: web services; quality of service; QoS prediction; service recommendation; real-time adaptability; approximate matrix multiplication.
Special Issue on: Security for Cloud Computing
Searchable Symmetric Encryption Based on the
Inner Product for Cloud Storage
by Jun Yang, Shujuan Li, Xiaodan Yan, Baihui Zhang, Baojiang Cui
Abstract: Searchable encryption enables the data owner to store their own data after
encrypting them in the cloud. Searchable encryption also allows the client to search over
the data without leaking any information about it. In this paper, we rst introduce a
searchable symmetric encryption scheme based on the inner product: it is more ecient
to compute the inner product of two vectors. In our construction, the parties can be Data
Owners, Clients or the Cloud Server. The three parties communicate with each other
through the inner product to achieve the goal that the client can search the data in the
cloud without leaking any information on the data the owner stored in the cloud. We then
perform a security analysis and performance evaluation, which show that our algorithm
and construction are secure and ecient.
Keywords: Searchable Encryption; Searchable Symmetric Encryption; Inner Product;
the Cloud Server; Security.
Lattice-based Searchable Public-key Encryption Scheme for Secure Cloud Storage
by Run Xie, Chunxiang Xu, Changlian He, Xiaojun Zhang
Abstract: With the popularity of cloud storage and the improvement of awareness of data privacy, the user's sensitive data is usually encrypted before uploading them to the cloud. Searchable encryption is a critical technique on promoting secure and efficient cloud storage. In particular, public key encryption with keyword search (PEKS) provides an elegant approach to achieve data retrieval in encrypted storage. However, all existing searchable public-key encryption schemes only provide the security based on classical cryptography hardness assumption. With the enhancement of cloud-computing power and the development of quantum computers, these schemes will be insecure. In this paper, we propose a new searchable public-key encryption scheme with a designated tester(dPEKS). Our scheme has notable advantages: Firstly, our scheme is the first searchable public-key encryption scheme based on lattice hardness assumptions. Currently, the lattice-based cryptography is considered to be secure even if quantum computers are ever developed. Therefore, our scheme is the promising candidate for traditional schemes. Secondly, our scheme achieves the trapdoor indistinguishability. The trapdoor indistinguishability implies the security against outside off-line keyword guessing attacks(KGAs). Until now, only few schemes can resist outside off-line KGA. In Boneh et als original framework, the inside keyword guessing attacks(KGAs) is considered inevitable. In this sense, our scheme provides the strongest security level. Lastly, our scheme can achieve the trapdoor anonymity for server.
Keywords: dPEKS ; searchable encryption; trapdoor indistinguishability; lattice; keyword-guessing attack; cloud storage.
Key-Aggregate Searchable Encryption under Multi-owner Setting for Group Data Sharing in the Cloud
by Tong Li, Zheli Liu, Chunfu Jia, Zhangjie Fu, Jin Li
Abstract: In recent years, the encryption with keyword search has been widely used
in cloud data sharing system to protect privacy and confidentiality
when the ciphertext is retrieving. However, selectively sharing encrypted
data and related searching abilities among different users via the
existing searchable encryption technology certainly will generate a
large number of searching trapdoors making the system inflexible and
impractical. In this paper, we propose the concept of ``multi-owner
key-aggregate searchable encryption'' scheme and its implementation,
in which a user can only submit a trapdoor for querying the documents
shared by multiple owners who only need to distribute an aggregate key for
sharing massive data. Thus, the scheme supports effective data sharing
for both multiple owners and users by reducing unnecessary trapdoors which
is hard for generating by mobile devices during the querying step.
Finally we conduct security analysis and performance evaluation
which can show that our system is practical and secure.
Keywords: cloud storage; searchable encryption; data sharing; key-aggregate.