Vol. 37 · No. 4

The popularity of cloud computing is increasing by leaps and bounds. To cope with resource demands of increasing number of cloud users, the cloud market players establish large sized data centers. The huge energy consumption by the data centers and liability of fulfilling Quality of Service (QoS) requirements of the end users have made resource allocation a challenging task. In this paper, energy and QoS aware resource allocation approach which employs Antlion optimization for allocation of resources to virtual machines (VMs) is proposed. It can operate in three modes, namely power aware, performance aware, and balanced mode. The proposed approach enhances energy efficiency of the cloud infrastructure by improving the utilization of resources while fulfilling QoS requirements of the end users. The proposed approach is implemented in CloudSim. The simulation results have shown improvement in QoS and energy efficiency of the cloud.

Sensor Clouds have opened new opportunities for agricultural monitoring. These infrastructures use Wireless Sensor Networks (WSNs) to collect data on-field and Cloud Computing services to store and process them. Among other applications of Sensor Clouds, frost prevention is of special interest among grapevine producers in the Province of Mendoza – Argentina, since frost is one of the main causes of economic loss in the province. Currently, there is a wide offer of public cloud services that can be used in order to process data collected by Sensor Clouds. Therefore, there is a need for tools to determine which instance is the most appropriate in terms of execution time and economic costs for running frost prediction applications in an isolated or cluster way. In this paper, we develop models to estimate the performance of different Amazon EC2 instances for processing frosts prediction applications. Finally, we obtain results that show which is the best instance for processing these applications.

Cost optimization for workflow scheduling while meeting deadline is one of the fundamental problems in utility computing. In this paper, a two-phase cost-efficient scheduling algorithm called critical chain is presented. The proposed algorithm uses the concept of slack time in both phases. The first phase is deadline distribution over all tasks existing in the workflow which is done considering critical path properties of workflow graphs. Critical chain uses slack time to iteratively select most critical sequence of tasks and then assigns sub-deadlines to those tasks. In the second phase named mapping step, it tries to allocate a server to each task considering task's sub-deadline. In the mapping step, slack time priority in selecting ready task is used to reduce deadline violation. Furthermore, the algorithm tries to locally optimize the computation and communication costs of sequential tasks exploiting dynamic programming. After proposing the scheduling algorithm, three measures for the superiority of a scheduling algorithm are introduced, and the proposed algorithm is compared with other existing algorithms considering the measures. Results obtained from simulating various systems show that the proposed algorithm outperforms four well-known existing workflow scheduling algorithms.

A key issue for full-time video surveillance is to search or establish a reference image of background which corresponds to current video frame. However, background that was ever in presence long time ago is enclosed or discarded due to background forgetting assumption. How to rapidly pick up or even rebuild long-term background needs to be discussed. This paper aims to present a framework for background maintenance in order to solve the problem. A piecewise memorizing framework is proposed for matching, updating and even rebuilding long-term background. Based on the metaphors of psychological selective attention theory, the framework is composed of a prior piecewise perception processor for intensity stationary test. Besides, a hierarchical memorizing mechanism constitutes the other part of the framework for overcoming the exponential forgetting of long period background appearances. Applied to Gaussian mixture model (GMM), this framework is capable of maintaining short-term background states, identifying long period background appearances, and rapidly adjusting to new background states according to different expressions derived from the prior perception of scene intensity changes. Its effectiveness can be demonstrated by experimental results for solving various typical problems.

In this paper we propose very efficient itemset representation for frequent itemset mining from transactional databases. The combinatorial number system is used to uniquely represent frequent k-itemset with just one integer value, for any k>=2. Experiments show that memory requirements can be reduced up to 300

Vertical partitioning is the process of subdividing the attributes of a relation into groups, creating fragments. It represents an effective way of improving performance in the database systems where a significant percentage of query processing time is spent on the full scans of tables. Most of proposed approaches for vertical partitioning in databases use a pairwise affinity to cluster the attributes of a given relation. The affinity measures the frequency of accessing simultaneously a pair of attributes. The attributes having high affinity are clustered together so as to create fragments containing a maximum of attributes with a strong connectivity. However, such fragments can directly and efficiently be achieved by the use of maximal frequent itemsets. This technique of knowledge engineering reflects better the closeness or affinity when more than two attributes are involved. The partitioning process can be done faster and more accurately with the help of such knowledge discovery technique of data mining. In this paper, an approach based on maximal frequent itemsets to vertical partitioning is proposed to efficiently search for an optimized solution by judiciously pruning the potential search space. Moreover, we propose an analytical cost model to evaluate the produced partitions. Experimental studies show that the cost of the partitioning process can be substantially reduced using only a limited set of potential fragments. They also demonstrate the effectiveness of our approach in partitioning small and large tables.

Two or more object-oriented components located in networked computers can form a distributed system to exchange information and execute methods. The most known approaches include object request broker architectures (e.g. CORBA), messaging-service architecture (e.g. based on ZMQ or JMS) or some variant of Service Oriented Architecture (e.g. SOAP). One of new approaches in the field is the OPC-UA protocol. While having common parts with all aforementioned architectures, it brings very rich and extensible information modelling capabilities, versatility and dynamic address space model, among others. This paper proposes a mapping of information model (applicable in the OPC-UA protocol) into class and object structure of an object-oriented programming language. Special attention is paid to whether given programming language is statically or dynamically typed, with examples and applications in C++ for the former case and Python for the latter. The study also covers the cases of using the proposed mapping at both server- and client-side of OPC-UA software.

The Web contains a large amount of structured tables, most of which lacks header rows. Algorithmic approaches have been proposed to recover semantics for web tables by annotating column labels and identifying subject columns. However, state-of-the-art technology is not yet able to provide satisfactory accuracy and recall. In this paper, we present a hybrid machine-crowdsourcing framework that leverages human intelligence to improve the performance of web table annotation. In this framework, machine-based algorithms are used to prompt human workers with candidate lists of concepts, while an improved K-means algorithm based on novel integrative distance is proposed to minimize the number of tuples posed to the crowd. In order to recommend the most related tasks for human workers and determine the final answers more accurately, an evaluation mechanism is also implemented based on Answer Credibility which measures the probability of a worker's intuitive answer being the final answer for a task. The results of extensive experiments conducted on real-world datasets show that our framework can significantly improve annotation accuracy and time efficiency for web tables, and our task reduction and answer evaluation mechanism is effective and efficient for improving answer quality.

In text categorization, a document is usually represented by a vector space model which can accomplish the classification task, but the model cannot deal with Chinese synonyms and polysemy phenomenon. This paper presents a novel approach which takes into account both the semantic and statistical information to improve the accuracy of text classification. The proposed approach computes semantic information based on HowNet and statistical information based on a kernel function with class-based weighting. According to our experimental results, the proposed approach could achieve state-of-the-art or competitive results as compared with traditional approaches such as the k-Nearest Neighbor (KNN), the Naive Bayes and deep learning models like convolutional networks.

Similarity method has been in science for several centuries. The basis of the study is closely connected with mathematical linguistics. This approach has allowed obtaining new results in the analytical geometry, which, in turn, is used in different applications in information technology. The results are described briefly. Binary relations in linguistics and geometry are compared with the position of system analysis. The modified hypothesis of space as a binary structure is put forward on the basis of singular linear transformations. The hypothesis of the human sensory system is given shortly. Architecture computing appliance for solving this class of problems is proposed. The modified method is also applied in pattern recognition. The presence of symmetry in natural languages is shown briefly.