ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES" N. 12, 2014

To the contents

A. V. Antonov, Dean of Faculty, e-mail: antonov@iate.obninsk.ru, N. L. Salnikov, Professor, M. O. Khromova, Graduate Student, V. A. Chepurko, Associate Professor Obninsk Institute for Nuclear Power Engineering (INPE MEPhI).

Justification of Kernel Estimation of the Failure Intensity of Restores Technical Systems

In this article, we propose a new method for estimating the reliability of recoverable objects. This paper continues previous work of these authors. The method is based on the kernel estimation failure intensity ungrouped, grouped flow failures and mixed data. The second stage is estimated by the renewal equation of density reduction. In this paper we prove an asymptotically unbiased estimates. The total flow is the sum ot flows of failures ot m similar systems. We study ehe limiting distribution of the t_k — k-th moment of the total flow of failures. For large enough m, the distribution of t_k is close to the normal distribution.
Keywords: renewal equation, renewal function, failure intensity, kernel estimate, the smoothing parameter

P. 3—8

A. Korobko, Researcher, T. Penkova, Researcher, e-mail: Penkova_t@icm.krasn.ru Institute of Computational Modelling SB RAS, Krasnoyarsk

The Integral OLAP-Model of the Domain for Analytical Decision Making Support

This paper presents the approach of analytical decision making support based on the integral OLAP-model of the domain. Authors propose a method of constructing the integral OLAP-model of the domain as a concept lattice of multidimensional cubes. This method consists of a formal concept analysis of OLAP objects based on expert knowledge of the structure of analyzing objects and their comparability and contains the following basic stages: determining the set of analyzed objects; forming the formal context; generating the formal cube-concepts; constructing the concept cube-lattice. The integral OLAP-model includes all possible combinations of facts and dimensions of the domain and allows the decision maker to discover the unexpected analytical dependencies of the set of all actual analyzed objects. The implementation of the integral OLAP-model for analytical decision making support is described in this paper. The realization of integral model improves the effectiveness of analytical decision making support based on OLAP.
Keywords: integral OLAP-model, on-line analytical processing, formal concept analysis, decision making support

P. 8—13

N. N. Svetushkov, Senior Researcher, e-mail: svt.n.n.@mail.ru, Moscow Aviation Institute (National Research University)

The Method in Multidimetnsional Unsteady Heat Conduction Problems

The article describes the developed computer software environment that allows to find the distribution of temperature fields for spatially extended geometrically complex parts made of composite materials under external heat. As the calculation method uses a new algorithm called "separated strings method", in which partial differential equations are reduced to an equivalent system of integral equations. Author has analyzed the shortcomings of existing calculation methods and numerical schemes. There are some advantages of the developed approach, including the possibility of efficient parallelization that is really relevant for the solution of heat conduction problems for spatially extended objects, which include parts of the wing and the fuselage of the aircraft. Integral description of heat transfer processes, not only allows a more efficient simulation of heat transition by suppressing the oscillatory behavior of the solution in the case of large temperature gradients, but also a natural way to take into account heterogeneity of different products, without the use of various "seaming " approaches. An important advantage of the algorithm is ability to assess the accuracy of the numerical solution with the residual function in the system of integral equations. The article shows typical temperature profiles as in the transient and quasi-stationary cases at various ways of reinforcing insulating materials. There is conclusion about the need explicit consideration of the internal structure of the composite material in the case of large heat loads that applies to parts made of composites in aerospace engineering.
Keywords: software, simulation, composite materials, numerical methods, the heat equation, integral equations, the accuracy of calculations

P. 14—19

M. G. Matveev, Professor, E. A. Sirota, Assosiate Professor, e-mail: sirota_ea@sc.vsu.ru, Voronezh State University

Development and Research of Statistical Models of Unsteady Multidimensional Time Series of Atmospheric Temperature Conditions Heterogeneity

In this paper we study, construction and analysis of models of vector autoregression (VAR) with variable parameters, continuously dependent on changes in the atmosphere, to describe the multi-dimensional non-stationary time series of temperature conditions heterogeneity statistics. In this paper we have considered a model of mean daily temperature changes in the nodes flat grid above the ground. On the set of basic statistical data in the vicinity of a given node is proposed procedure for constructing homogeneous classes on the basis of statistics and postulating uniformity of physical conditions and verification of these conditions on measurable criteria. Next to these classes of homogeneous statistics carried out the construction of models VAR temperatures and conducted according to the approximation of the model parameters from the VAR state of the atmosphere. The main result of this study can be considered an experimental study the possibility of using this model to analyze the dynamics of unsteady processes in meteorology.
Keywords: multi-dimensional non-stationary time series, meteorology, vector autoregression, temperature trend, modeling classes homogeneous statistic

P. 20—24

R. E. Asratian, Leading Researcher, e-mail: rea@ipu.ru, Institute of Control science RAS

Distributed Processing of Data Queries in Multi-Server Network Environment

The principles of the organization of the network protocol providing possibility of distributed processing of one query in a chain of several servers are considered. Described approach represents "network" generalization of the known mechanism of "pipe" and is focused on an exchange of messages between programs running on different network nodes on the basis of standard input and output datasets of processing tasks.
Keywords: distributed systems, Web-technologies, Internet technologies, remote interactions

P. 25—30

Yu. I. Shokin¹, Director, e-mail: dir@ictnsc.ru, A. Yu. Vesnin², Head of a Laboratory, e-mail: vesnin@math.nsc.ru, A. A. Dobrynin², Senior Researcher, e-mail: dobr@math.nsc.ru, O. A. Klimenko¹, Senior Researcher, e-mail: klimenko@ict.nsc.ru, E. V. Rychkova¹, Researcher, e-mail: helen@ict.nsc.ru
¹Institute of computational technologies of the Siberian Branch of the Russian Academ
²Sobolev Institute of Mathematics of the Siverian Branch of the Russian Academy of Sciences

Analysis of a Web-Space of Academic Communities by Method of Webometries and Graph Theory

Web-spaces of two famous academic communities in Russia and in Germany, the Siberian Branch of Russian Academy of Sciences and the Fraunhofer-Gesellschaft, are investigated by webometries and graph theory methods. The content of sites and links between them are analyzed by search engine services and special programs. Webometries methods are applied for statistical analysis of content of scientific organizations sites and ranking sites for each of communities. For studying of information interaction between sites, the graph representation of the web-space is used. Vertices of such a web-graph correspond to sites and a relation between two vertices is defined by hyperlinks from one site to another. For both communities structural and metric properties of web-graphs and their fragments are studied. It admitted to compare their numerical and structural characteristics. The similarities and differences between web-spaces of the communities are founded. The obtained results can be used for a future improving of academic web communication.
Keywords: web-space, webometries, graph theory, strongly connected component, graph, graph clique

P. 31—40

M. B. Bulakina, Associate Professor, Moscow State University of Economics, Statistics and Informatics, e-mail: bulakina@smail.com, A. D. Ivannikov, Professor, Deputy Director, Institute for Design Problems in Microelectronics, e-mail: adi@ippm.ru

Federal Portal "All-Russian Education" as a Mean for Educational Information Resources Aggregation

Design and Development of the Federal Portal "All-Russian Education" is described. The Portal is considered as an aggregation point for the information posted on the Internet in the field of education, in particular, electronic educational resources. The Portal was created as an integrating element for the educational portals system. It has various services and educational databases. One of the central elements is educational electronic resources catalog containing metadata for resources for different subjects and different educational levels. Statistical analysis of portal audience is described and analyzed.
Keywords: portal, educational resources, data aggregation, the system of federal portals

P. 41—47

A. V. Maltsev, Senior Researcher, e-mail: avmaltcev@mail.ru, Scientific Research Institute of System Analysis (Russian Academy of Sciences)

The Formation Technology of Earth's Detailed Surface Texture for Space Training Complexes

To achieve high realism when synthesizing images of Earth's three-dimensional virtual model for space simulation-training complexes of cosmonaut training, highly detailed texture of Earth's surface must be used in visualization process. At this work the technology are proposed for generation such texture from initial photographic image set of Earth's small plots of same size (also called "tiles") to further application this texture when Earth's models are rendered in visualization systems using ultra-large texture technology. Considered approaches are based on using distributed computing on GPU with CUDA architecture. Furthermore, in this paper the software complex that has been created based on proposed decisions is described. This complex consists of two applications for Microsoft Windows operating system and one script for the program Adobe Photoshop. They together allow user to prepare required texture of Earth's surface by few steps.
Keywords: visualization, video trainers, Earth's surface textures, parallel computing, graphics processors

P. 48—53

Π. Yu. Timokhin, Research Associate, e-mail: webpismo@yahoo.de, M. A. Torgashev, Head of Sector

Real-Time Visualization of Space Scenes in High Dynamic Range

The article considers the problem of real-time visualization of space scenes. In the absence of atmosphere in space, light has a very tough character and images, observed by an astronaut on orbit, have a huge dynamic range. One of the most effective approaches to solving the problem of adequate environment simulation is synthesis of virtual scene in the high dynamic range, followed by tonal mapping to a range of display. The study proposes a new technology to solve this problem in real-time using parallel computing on the GPU. There are considered in detail the questions of calculating the average and the maximum frame's luminance, simulating the effect of "bloom" and mapping images to the low dynamic range using a modified version of the Reinhard's global tone mapping operator. This paper presents a new approach to calculating the average luminance which has a number of advantages over existing developments, as well as efficient algorithm for computing global luminance parameters of the frame by means of geometry shader.
Keywords: visualization, high dynamic range, tonal mapping operator, space training systems, parallel computing

P. 53—60

V. K. Ignatovich¹, Leading Researcher, e-mail: v.ignatovi@gmail.com, I. M. Salamatin¹, Senior engineer, e-mail: salam@nf.jinr.ru, K. M. Salamatin², Graduate Student, e-mail: del@xaker.ru, A. E. Senner², Assistant Professor, e-mail: del@tmpk.ru
¹Joint Institute for Nuclear Research, Frank Laboratory of Neutron Physics,
²International University "Dubna"

Automation of Experiments in the Field of Neutron Spectrometry Using Network Technologies

A new structure of a distributed software for experiments automation system (EAS) is proposed. The EAS includes: 1) a subsystem describing the experimental procedure (SDEP); 2) the experiment control program (ECP); 3) distributed components messaging environment (DICME); 4) components for sample environment at the instrument cjntrol; 5) Data Acquisition (DAQ) subsystem; 6) components for auxiliary operations. SDEP contains data base (DB) with documentation of controllers, devices connected to them and program components, and two dialog programs: 1) a program producing documentation of components; 2) program of preparation of experiment description as a lisn jof EAS states. ECP program receives from SDEP the list of EAS states, selects the description of the next state of the system (list of the data registration conditions) and sends a description of each condition to the intermediary DiCME. In each state of EAS data registration is completed. Mediator DiCME, developed using network technology, automatically searches and dynamically binds components. It provides an interface to component's procedures independent on the computer network address.
As a result, the components implemented in different experiments and in different EAS can be used without modification. The presented approach was applied in the development of EAS software for several instruments at neutron sources IBR-2 and IRENE in JINR. It can be also applied in other areas, for example, for technological process automation.
Keywords: experiment automation, dynamic coupling, asynchronous communications, heterogeneous distributed systems, programming technology

P. 61—66

A. S. Borovsky, Deputy Director, e-mail: borovski@mail.ru, The Institute of risk management and comprehensive security — the Orenburg state agrarian University

The Method of Determining the Optimal Level of Capabilities Engineering Protection on the Object in the Task of Designing Physical Protection Systems

A method for optimal placement of funds engineering protection on structural and logical object model and determine the optimal level of their capabilities (detection delay).
Presents the object model, in the form of two levels: the structural-logical representing the object structure and the relationships between its elements and functional, as in the object model, you must consider the data about the production process of the object.
The notion of control points, which allows to abstract from the physical nature of the source of the event, and to operate only logical elements. Physically, the point of control is one or more elements of the engineering and technical means of protection, i.e. a specific physical device, such as a signaling device (sensor), the actuator (electromagnetic lock), or device that controls access.
The model of the physical protection system are presented in the form of a set of control points and determined the minimum required their part to protect one critical element of the object.
Introduced the following notion of control points: point detection; access point; the point of the video; the point of delay.
Each control point is characterized by two parameters: the probability of detection and probability of delay, which determine the level of capabilities of the engineering and technical protection.
Defined standard genetic algorithm is responsible for the quality of the solution (optimization) and task data adapted to the subject area of research.
The novelty of this work is to use the model of the logical inspection points that are not associated with specific physical devices ITSO, which significantly simplifies the algorithm finds the optimal solution.
Keywords: the system of physical protection, engineering and technical means of protection, the probability of detection, probability of delay, the algorithm of the search paths, the uncertainty, the standard genetic algorithm, fuzzy variables

P. 67—75

To the contents