Earticle

This paper tries to address the problem of deriving the different features of a system and then having a way of making informed decisions about them based on their level of importance to the whole system as well as to each other depending on several given factors. The use of Systems Features Analysis (SFA) to derive the features and Applied Hierarchy Process (AHP) to decide on their importance fits the given situation and they are described in this paper. These tools are successfully applied to two system development cases, a whole system and some components of a system respectively, which showed their effectiveness and usefulness. An AHP-based software called SuperDecisions is utilized to immediately use AHP in the software design and development process in the shortest possible time.

Information Systems are becoming more complex as time passes and this complexity needs to be managed. Moreover, the complete matching between applications requisites and delivered software systems is one of the most important elements in software development, since cost effectiveness, efficiency and customer satisfaction are the key of success. In order to achieve these goals, many designers and analysts are applying Model-driven Engineering and UML modelling to their projects, which is gaining an important relevance in the Information Technology environment. This article proposes an extension to the standard UML metamodel with the scope to improve the development process and to grant a high quality of the software delivered to the customers. The metamodel extension contains a collection of elements, with a proper semantic and a set of rules to link them in order to create robust and expressive models, which can be transformed almost seamlessly into code with the help of a Java framework developed in conjunction with the custom UML profile.

Distributed computing systems have evolved over decades to support various types of scientific applications and overall computing paradigms have been categorized into HTC (High-Throughput Computing) to support bags of tasks which are usually long running, HPC (High-Performance Computing) for processing tightly-coupled communication-intensive tasks on top of dedicated clusters of workstations or Supercomputers, and Data-intensive Computing leveraging distributed storage systems and parallel processing frameworks. Many-Task Computing (MTC) aims to bridge the gap between traditional HTC and HPC by building efficient middleware systems throughout employing lightweight task dispatching mechanisms, minimizing data movements, leveraging data-aware scheduling, and proposing of next generation Exascale storage systems. Recent emerging applications requiring millions or even billions of tasks to be processed with relatively short per task execution times have driven this new computing paradigm. In this paper, we investigate concepts and technologies of MTC and propose guidelines for building an efficient and effective middleware system to fully support MTC applications. Throughout our short survey about challenges, systems and applications of MTC, we argue that a next generation distributed middleware system must effectively leverage distributed file systems, parallel processing frameworks, decentralized data/compute management systems, and dynamic load balancing techniques to solve the most challenging and complex scientific problems.

Most model transformation approaches expressed to transform the static model structures of a system, but not involved with the behavioral model. Our previous model transformation [10] also focused on the structural model, especially class diagram, which was restricted to generate a detailed code. To solve this problem, we propose model transformation with both structural and behavioral models, that is, a message sequence diagram with a class diagram for developing heterogeneous software. This approach makes it possible to generate detailed codes through the static & behavioral expression of a system. This is also better to transform more specific design than the previous structure-based model transformation based on only static structure [11-15]. We show the application cases to perform model transformation on the three platforms.Android, iPhone, and Windows Phone.

This paper proposes a semi-parametric software reliability model (SRM) based on a mixed gamma distribution, so-called the mixed gamma SRM. In addition, we develop the param- eter estimation method for the mixed gamma SRM. Concretely, the estimation method is based on the Bayesian estimation and the parameter estimation algorithm is described by MCMC (Markov chain Monte Carlo) method with grouped data. In numerical experiment, we compare the mixed gamma SRM with several parametric SRMs, and examine the per- formance on the tting ability of proposed model.

Ever since their introduction, social networks have attracted millions of users and play very significant and important role in world’s communication system. Social networks are now becoming a useful to develop business, social and educational spheres in the country, but also can have their own limitations and problems. The primary goal of this research is to reveal attitudes of Kazakhstan social network users to worldwide and domestic resources, their preferences, and to investigate the question of further domestic networks development. Current position of domestic Kazakhstan social networks is weak and requires improvements. Further integration and development can bring benefits to businesses, educational sphere and to the whole society. Benefits of strong social network extend to the level of individual person, giving new opportunities and options. This study is expected to make contributions for Kazakhstan based social network services (SNS) to explore potential opportunity to extend their services and enrich policies and practices in implementing theirs businesses.

Ranch industry has grown bigger. In Australia, ranches have very large number of live-stock commodities: cattle, lambs, and muttons. To manage such a very large scale commodi-ties, they need to install sensor network with MapReduce of Hadoop; since the sensor net-work generates a huge amount of data. The ranch is divided into several patterned regions and a lot of hubs are installed in there for retrieving the sensor data. However, when the sen-sor moves to an overlapped area among some hubs, the sensor data are transmitted to all hubs covering the area. Obviously, these data is redundant. Therefore, we propose removal technique to delete the redundant data and to efficiently process the data on the map phase. In order to detect redundancy, the data will be compared using some parameters, and then the detected redundant data will be deleted according to some rules.

In this paper, we introduce the design of fuzzy respective space-based neuro-fuzzy networks for pattern recognition. The proposed networks are realized by partitioning of the fuzzy respective input space to generate the fuzzy rules. The respectively partitioned spaces using fuzzy respective input space express the rules of the networks. The consequence part of the rules is represented by polynomial functions. The coefficients of consequence part of the rules are learned by the back-propagation algorithm. And we also optimize the proposed networks using real-coded genetic algorithms. A numerical example is given to evaluate the validity of the proposed networks for pattern recognition. As a result, this paper shows that the proposed networks have the good result together with fewer rules.

We handle a problem to assign the starting offsets of a collision-free schedule in a certain form called a well-formed gcd notation. Each activity in an activity set recurs with its own period. If more than one activity occurs at the same time slot, it is a collision. If a schedule can be constructed as a well-formed gcd notation, the schedule is known to be collision-free. In this paper, we introduce a well-formed gcd notation and present an algorithm to assign the starting offsets of a collision-free schedule. We also propose a restricted form of a sufficient condition for detecting a collision-free periodic schedule, which can be decided in a tractable time constraint.