ISSN 0236-235X (P)
ISSN 2311-2735 (E)

Journal influence

Higher Attestation Commission (VAK) - К1 quartile
Russian Science Citation Index (RSCI)

Bookmark

Next issue

4
Publication date:
09 September 2024

Journal articles №4 2015

31. Representativeness of Intel Sandy Bridge performance counters based metrics for memory access latency analysis [№4 за 2015 год]
Authors: Ivanov E.Yu. ( i@eivanov.com, eiva@tbricks.com) - The National Research University of Information Technologies, Mechanics and Optics, Tbricks AB; Kosyakov M.S. (mkosyakov@gmail.com, mkosyakov@tbricks.com) - The National Research University of Information Technologies, Mechanics and Optics, Tbricks AB, Ph.D;
Abstract: This paper examines the representativeness of metrics based on Intel Sandy Bridge performance event counters and memory access latencies for data processed in RAM. It contains detailed overview and analyses of frequently used metrics recommended by Intel documentation. The authors have implemented a synthetic test application, which allows measuring memory access latencies for serial, per-page and random memory accesses and different sizes of working set and its elements processed by test application. The paper describes an approach to access performance event counters based on Linux perf utility. The test presented memory access latencies and calculated metrics based on performance event counters. The experimental studies showed that CPI, SCPI and PSRC metrics recommended by Intel and cache-miss ratio with serial memory access have linear dependency with measured memory latencies. Besides, the metrics values change proportionally to the change of memory latencies with the coefficient close to one. The random access results showed that most efficient metric for performance is the number of cycles wasted on bus waiting for data, because it always has high representativeness of memory access latencies and does not depend on data location. According to authors’ analyses, this metric is the most useful to evaluate memory access performance optimization efficiency. The results showed that the metrics considered in this paper can be used to analyze memory characteristics of application, which process data in RAM.
Keywords: intel sandy bridge, cache-memory, speed, optimisation, profiling, memory access, memory access latencies, event processor, performance event counters
Visitors: 10279

32. Data replication in hierarchical information systems with unstable node connectivity [№4 за 2015 год]
Authors: Sorokin V.E. (sorokinve@yandex.ru) - Developed in the Research Institute "CENTERPROGRAMMSYSTEM", Ph.D;
Abstract: A particular feature of many hierarchical information systems (IS) with relational databases (DB) in their nodes as the core is impermanent communication between the nodes. It prevents classical use of a centralized or distributed database and assigns additional special message exchange and data consolidation functions including external sources to special software, which solves system functional tasks. Replication of required fragments of other IS nodes DB in a local DB consolidates with information from external sources. It allows to simplify and increase harmonization of various IS nodes software. The article proposes an approach to original DB structure transformation based on creating separate for different data sources inherited from source tables. The approach is adapted to asynchronous “master/slave” replication of hierarchically related DB. The paper shows the possibility of using flags and procedures managed by triggers to implement various strategies for data consolidation, as well as a simple adaptation of the original software to solve IS functional problems on the converted DB. There is an example of the described approach implementation for widespread and used in many domestic certified software tools of an object-relational database management system PostgreSQL with an asynchronous “master/slave” replication system Slony-I. There is a possibility to overcome limitations of an asynchronous “master/slave” replication system using dblink support modules of session connection to other DB and postgres_fdw foreign-data wrapper, as well as replication system joint using with these modules on the converted DB.
Keywords: external data wrapper, inheritance tables, asynchronous replication, relational database, hierarchical information system
Visitors: 9014

33. Complex industrial projects data representation models in automated information systems of industrial institutions [№4 за 2015 год]
Authors: Dli M.I. (midli@mail.ru) - (Smolensk Branch of the Moscow Power Engineering Institute, Ph.D; Stoyanova O.V. (ovstoyanova@list.ru) - Smolensk Branch of the Moscow Power Engineering Institute, Ph.D; Belozersky А.Yu. (ovstoyanova@ list.ru) - D. Mendeleev University of Chemical Technology of Russian Federation (Professor), Ph.D;
Abstract: A great number of various information systems is used during the process of complex industrial projects management. Each of them solves independent projects data processing tasks. Thus, as for new production setup projects, product information is distributed between computer-aided design systems (CAD) and automated process control systems (APCS), information about necessary resources is distributed between APCS, management information system (MIS), automated project management system (APMS), etc. Project data exchange between stated systems is complicated by using different hardware and software, data formats, data representation models. The article considers object and relation data representation models of complex industrial projects. The first type of model is used as base of data processing using object-oriented programming languages. The relation model is necessary for data storage in industrial institutions databases. This situation is caused by information technologies evolution, i.e. data processing a pplication development using object-oriented programming and relation DBMS dominance. Different units of information systems use different data models. This requires decision-making on transforming data representations. There is no one-stop solution for the stated task for now. The paper considers the example of such transformation for an object model in Java and for a relation model in DBMS Oracle.
Keywords: automated information system, project management, project data model, industrial project
Visitors: 14082

34. Intelligent information system for retail demand forecasting and inventory optimization [№4 за 2015 год]
Author: Gritsay A.A. (allgrit@forecastnow.ru) - Tver State University;
Abstract: The paper presents inventory management and stock optimization system for retail trade. The system is focused on small and medium retailers and wholesale trade customers. The estimated savings make a total of 20–30 % of the stock. System can increase satisfied demand by 5–7 %. The article describes the main features of the system: data preprocessing, demand forecasting, calculation of the optimal service level, calculation of optimal inventory, order calculation for a supplier. Data preprocessing cleans data from sharp emissions, marketing campaigns impact, holiday impact and the lack of goods in the warehouse. Demand forecasting is based on the preprocessed time series of the sales. The characteristics of sales history time series are calculated for each product. The prediction algorithm is selected according to the characteristics. Given service level defines the reserve stock. The optimal service level is calculated according to the strategy of trading company investment return maximization. Optimal inventory is calculated based on the calculated service level and the chosen forecasting algorithm. Available goods in stock, goods in transit, as well as provisions for individual customers determine the optimal order size of the item. After that row order restrictions are applied (e.g. packaging multiplicity), and then order restrictions to a supplier as a whole (e.g. weight and dimensional). The paper provides an example of applying the system and the conclusion about economic effect from its implementation in a real trading enterprise.
Keywords: order to a supplier, optimal service level, demand forecasting, neural network, inventory management system, inventory stock optimization, inventory management
Visitors: 9533

35. Intelligent information system for rolling stock carriage equipment faillures forecasting on railways [№4 за 2015 год]
Author: Ivanova E.I. (kkateivanova@gmail.com) - Tver State University;
Abstract: The “Strategy of railway engineering development until 2030” presents rail transport increased safety as one of the most important problems for research. One of the main goals of JSC “Russian Railways” in this strategy is creation of complex systems for diagnostics and forecasting of rolling stock infrastructure technical condition. Analysis of publications on the subject shows that there is no information on using such systems on rolling stock on the Russian market. The article describes one approach to solve the abovementioned problem, which is using an intelligent information system focused on solving problems of equipment fault forecasting in rail transport. The article describes the main scientific results of such system development, the system architecture and its features, algorithms and implementation means, results and analysis of using this system on real data of rolling stock cars of JSC “FIC”. The described system is an adaptable and versatile in many ways. It is applied as a forecasting system to passenger cars, but in the future it can be adjusted and applied to other rail transport elements (freight cars, locomotives, electric, etc.). Its subsystems are independent and it is possible to try different algorithms within its subsystems without affecting the system as a whole. This system is realized with support from Russian Foundation for Basic Research within the competition of oriented basic research on important cross-disciplinary subjects on behalf of JSC “Russian Railways”. The developed system is tested on the car resources of JSC “TCW” equipped with the latest SKDU system of JSC “FIC” rolling stocks.
Keywords: echo state network, neural network, generic algorithm, soft computing, decision making, forecasting, diagnostics, rail transport, information system
Visitors: 12770

36. Algorithms and models for maintenance process automated control systems [№4 за 2015 год]
Authors: Burdo G.B. (gbtms@yandex.ru) - Tver State Technical University, Ph.D; Fedotova А.V. (afedotova@acm.org) - Bauman Moscow State Technical University (Associate Professor);
Abstract: Production systems of enterprises, which produce high technology products, often relate to multiproduct production. When using the technological processes automated control systems (TPACS) in multiproduct production, the main problems are caused by the high situation dynamism in the companies. This is due to the regular emergence of new contracts during a calendar year, resulting in production plans adjustment; the technical regulation approximation because of lack of time for technological preparation leading to operational planning inaccuracy. According to the authors, the proposed process automated control system (for example, aircraft technical maintenance) is devoid of these shortcomings. This is due to the following factors. The hierarchical structure of TPACS reflects the hierarchy of the enterprise production plans and is presented by subsystems of volume, calendar, operational planning and scheduling. TPACS as a system in whole is connected to an organization management system through information and is its subsystem. To ensure scheduling, there is information interaction with production departments. All the subsystems of TPACS are connected, so it is possible to correct all types of plans quickly. Planning is accomplished by using iterative procedures, which provide “thin” fulfillment of production plans. Therefore, higher level production plans adjusting is possible due to feedbacks, and this ensures their feasibility. When production planning, local capacity reserves and operating time funds are planned, the ways of using their reserves are suggested. The operational planning subsystem results in scheduling of production units. It is based on the relative priority method. The paper shows the refined statement of the production unit schedule problem. The particular features are associated with the performance of work by work groups and with the use of priority service schemes. The paper identifies the methods of forming the priority object service schemes that provide a variety of calculations and a choice of optimum in accordance with the established criterion. When refining, TPACS can be implemented as a decision support system in various diversified engineering enterprises, which provide service, repair and high technology products assembly.
Keywords: decision support, artificial intelligence, high-tech products, dispatching, planning, technological processes automated control systems
Visitors: 11612

37. The software design package for technological pipelines heating by heat tracers in computer-aided design systems [№4 за 2015 год]
Authors: Kokhov T.A. (tkohov@ggo.ru) - Giprogazoochistka; Korelshteyn L.B. ( korelstein@truboprovod.ru) - NTP Truboprovod (Deputy Director), Ph.D; Gartman T.N. ( gartman@muctr.ru) - D. Mendeleev University of Chemical Technology of Russian Federation (Professor, Head of Chair), Ph.D;
Abstract: Industrial plants use the following types of technological pipelines heating: trace heating; steam tracing; electric trace heating. Depending on performance specification or environmental conditions one of these types of heating can dominate in the project. About 10–15 % of design works (in wiring design) are technological pipelines heating design. In certain cases almost 40 % of pipelines in the design project should be heated by heat tracers. Sometimes this index can reach 50 %. Thus, about 80 % of heating engineering labor costs is elaboration of heat tracers’ traces in a 3D model. To speed-up engineering of technological pipelines heating by heat tracers the authors developed a software design package for technological pipelines heating implemented in 3D-design AVEVA PDMS. A distinguishing feature of the developed software design package for technological pipelines heating in comparison with other systems is that it is integrated into AVEVA PDMS. It allows executing modeling of technological pipelines heat tracers in 3D models and releasing necessary target documents: isometric drawings of pipelines. Software package functionality is verified when developing technological pipelines water heating by heat tracers of sulfur recovery.
Keywords: pipe manifold, perekidka, reversing, power supply, thermal satellite
Visitors: 10135

38. The approach to modeling, scaling-up and optimization of bioreactors based on computational fluid dynamics [№4 за 2015 год]
Authors: Guseva E.V. (eguseva@rally-online.ru) - D. Mendeleev University of Chemical Technology of Russian Federation (Associate Professor), Ph.D; Safarov R.R. (safarov@actremed.ru) - “Vozrozhdenie” Investment construction company; Menshutina N.V. (chemcom@muctr.ru) - D. Mendeleev University of Chemical Technology of Russian Federation, Ph.D; Boudrant J. (eguseva@rally-online.ru) - Universite de Lorraine, Laboratory Reaction and Process Engineering, LRGP, ENSAIA;
Abstract: During last 20 years computational fluid dynamics (CFD) proved itself as an instrument to solve problems with one or more phenomena, such as liquid flow, heat and mass transfer, chemical reaction. This article is devoted to the general approach to modeling, scaling-up and optimization of bioreactors using CFD. One of the steps the proposed approach is based on PIV (particle image velocimetry), which helps to determine particles velocities from their images. As an example the paper considers the process of modeling, scaling-up and optimization of lactic acid production by Lactobacillus casei bacteria in a stirred-tank batch bioreactor. The mechanical stress aspect (agitation rate) which influences on bacteria viability and productivity has been taken into account during the calculation. The maximum dissipation rate has been selected as a characteristic of mechanical stress. The authors defined the sizes of turbulence characteristic eddies in bioreactor that negatively affected bacteria. The authors have introduced the concept of stress zone specific volume, which has been used to define the limiting agitation rate in this bioreactor. The approach can be used for scaling-up and searching of optimal operation conditions.
Keywords: stress zone, scale, optimisation, cfd, bioreactor, microorganisms, cells/bacteria, mechanical stress, energy dissipation rate, modeling
Visitors: 12337

39. Halstead-technique based quality assessment framework for academic software products [№4 за 2015 год]
Authors: Afanasova A.I. (nastyaafanasova@yandex.ru) - Ulyanovsk State University (unior Researcher), ;
Abstract: Academic software products (ASP) assessment has a major impact on the acquisition level of complex computer and information disciplines. However, programmetry literature does not make the difference between ASP and professional software products (PSP). ASP projects, alternatively to PSP, do not have immediate achievement-oriented feedback from the market environment. Taking into account the scope of documentation, ASP frontal assessment comes out to be a labor-intensive work without necessary effect, which is a quick or visible accruement in student competency on account of unavoidable instructor’s subjectivity stemming from reliance on their personal experience, knowledge and beliefs. According to the authors, a solution is the creation of taskoriented methods, which are different from the PSP assessment methods, and in the development of specific instructor’s tools for ASP assessment. Due to the created non-biased feedback, such tools can contribute to students’ behavior modification. This will take the form of increased not only ASP quality, but also the quality of the entire project-based education process. The paper presents the development results of such ASP assessment tool. Based on the Halstead technique, an algorithm to compute quantitative performance indices for the ASP created by students in C++ is set forth. The authors suggest an original computer program written in C# as a framework tool. The expected results are supported by the product beta testing and trial operation of the assessment framework.
Keywords: с++, projectbased learning, halstead metrics, programmetry, student developed software, software quality assessment
Visitors: 6766

← Preview | 1 | 2 | 3 | 4