«Systems and Means of Informatics»
Scientific journal
Volume 24, Issue 2, 2014
Content | About Authors
Abstract and Keywords.
METHODS AND TOOLS FOR OPTIMAL PLANNING OF PROCESS PARAMETERS IN AFTERSALE SERVICE SYSTEMS OF HIGH TECHNOLOGY PRODUCTS
- I. N. Sinitsyn
- A. S. Shalamov
- I. V. Sergeev
- E. R. Korepanov
- V. V. Belousov
- T. S. Gumnikova
- V. S. Shorgin
- E. S. Agafonov
Abstract: The methods and software tools of aftersale systems optimization are considered based on cost criteria for a given level of technical operability coefficient are considered. A brief overview of current approaches to product lifecycle management of high technology products, based on theWestern concept of CALS (Continuous Acquisition and Life cycle Support) and its Russian analogue - IAS (integrated information support of products) is provided. A new stochastic approach to modeling of aftersale systems for products of high technology is presented. This approach is used to solve the actual problem of optimal design and operation of service system based on cost criteria. The methods for finding optimal values of service systems using different measures of efficiency are described. Methodological and instrumental software tools for cost optimization of spare parts supply and repairs for years to come, until the write-off of the device under restrictions on allocated financial resources and the need to maintain a given level of technical operability coefficient, are demonstrated.
Keywords: aftersale service system; high technology product; repair and supply of spare parts; level of technical operability of manufactured product park; cost of repair and supplies; limited budget; repair and supply of spare parts program optimization; software tool
RECONCILIATION OF AGGREGATED AND DISAGGREGATED TIME SERIES FORECASTS IN NONPARAMETRIC FORECASTING PROBLEMS.
- M. M. Stenina
- V. V. Strijov
Abstract: In many applications, there are the problems of forecasting a lot of time series with hierarchical structure. It is needed to reconcile forecasts across the hierarchy. In this paper, a new algorithm of reconciliation of hierarchical time series forecasts is proposed. This algorithmis based on solving the optimization problem with constraints. The proposed algorithm allows to reconcile the forecasts with nonplanar hierarchical structure and to take into account physical constraints of forecasted values such as nonnegativeness or maximal value. The algorithm performance is illustrated by the railroad stations occupancy data in the Omsk region. The quality of forecasts is compared with the quality of forecasts made by the optimal algorithm of reconciliation. Also, the algorithm performance is demonstrated for the nonplanar hierarchical structure of time series.
Keywords: hierarchical time series; nonparametric forecasting; empirical distri- bution; forecasts reconciliation
ESTIMATION OF THE EFFECTIVE BANDWIDTH OF A NODE IN AN INFO-COMMUNICATION TANDEM NETWORK.
- A. V. Borodina
- Е. V. Morozov
Abstract: The properties of the effective bandwidth (EB) regenerative estimate of a communication node in the tandem network are investigated. This problem has been studied earlier for a separate node with regenerative input. This setting is natural for acyclic networks because the input renewal process becomes positive recurrent regenerative while crossing the nodes of such a network (under the steady-state condition). Based on the theory of large deviations results, an approximation of the EB is proposed which quality is verified by simulation of a few tandem networks. The Weibull distribution with a light tail, the truncated Pareto distribution, and the exponential distribution are used for service time and the arrived workload, and the number of the nodes is varied from 2 to 40. It is shown that the EB estimator obtained by this approximation ensures the following condition: the overflow probability estimate is always less than the required given value (guarantee of quality of service). This result indicates the possibility to use the proposed approximation for choosing the EB values in the nodes in info-communication highly reliable tandem networks.
Keywords: tandem network; effective bandwidth; regenerative input; quality of service; theory of large deviations; approximation; statistical estimation; simulation
SYSTEM VERIFICATION TOOLS FOR RECURRENT SIGNAL PROCESSOR.
- V. S. Petrukhin
- D. Y. Stepchenkov
- N. V. Morozov
- Y. A. Stepchenkov
Abstract: A procedure of selecting and developing a software and hardware suite is studied which is intended for designing and debugging a nontraditional digital signal processor based on the recurrently-dynamic dataflow architecture | the recurrent signal processor (RSP). The experimental character of the developed RSP's architecture as well as a necessity for a finished master processor have predetermined Cyclone V family FPGA (Field Programmable Gate Array) (Altera) as the base of RSP implementation and Quartus II design software for its development. The powerful verification tools contained by Quartus II allow both reducing the time of obtaining the finished design and reducing the hardware expenses essentially. On the basis of the comparative analysis and selected criteria, the tools composition for debugging RSP is determined, and an optimal structure of hardware for debugging RSP is offered that allow for essential simplification of the verification process and for debugging RSP in a real hardware environment.
Keywords: debugging means; data flow architecture; verification
HIGH-PERFORMANCE LOAD GENERATOR FOR HIGH-FREQUENCY TRADING SYSTEMS VERIFICATION.
- D. K. Guriev
- M. A. Gai
- I. L. Itkin
- A. A. Terentiev
Abstract: The growing volume of orders generated by HFT (high-frequency trading) systems has posed the challenge of conducting exchange and brokerage systems testing in production-like environments. Specialized testing tools are used to ensure quality of high-load trading systems with high availability. The main requirement for such tools is that they should be capable of creating realistic, high loads using limited hardware infrastructure. The article describes a load injection tool developed for testing automated trading systems and an approach that ensures high performance. The tool is used when throughput and response times of large-scale exchange and brokerage platforms, the backbone of the technological infrastructure of financial markets, are measured.
Keywords: test automation; trading systems; HiVAT
USAGE OF PASSIVE TESTING TOOLS FOR CERTIFICATION OF TRADING SYSTEMS CLIENTS.
- A. N. Alexeenko
- A. A. Averina
- D. S. Sharov
- P. A. Protsenko
- I. L. Itkin
Abstract: The life cycle of exchange and brokerage software development, along with verification of functional and technical characteristics of the system, includes a mandatory stage of integration testing called clients certification. The stage is designed to ensure the compatibility of automated trading systems which are connected by means of financial protocols (such as FIX/FAST, ITCH, or specialized binary access interfaces) to an exchange or a broker. The article presents a bespoke tool developed to verify the compatibility of trading systems. The distinctive feature of the tool is a unified way of supporting multiple protocols. The article also provides a few examples of using the tool in self-certification of trading participants and during large-scale migrations of trading platforms.
Keywords: financial protocols; FIX-protocol, compatibility testing, self- certification; trading broker; stock exchange
SOURCE CODE AND PARTIAL SPECIFICATIONS ANALYSIS FOR AUTOMATED GENERATION OF UNIT TESTS.
- A. Andrianova
- V. Itsykson
Abstract: Low level of software quality is one of the main problems of software engineering. Automated testing is one of the most effective strategies used to improve the quality of the software. This paper describes a technique of automated unit tests creation combining both functional and structural approaches of software testing. In thismethod, information extracted fromthe original program is used to ensure test coverage of program paths. Partial specifications given in the form of contracts are used to form test oracles and to distribute parameters of tests on definition domain. The developed approach was implemented as a tool that analyzes Java programs and generates test cases for class methods in JUnit format, using CoFoJa to specify the contracts. Designed tool testing on a number of test cases showed efficiency of the approach.
Keywords: automated software testing; unit test generating; contract-based programming; code analysis; SMT-solver
DATA RACE DETECTION IN JAVA PROGRAMS USING SYNCHRONIZATION CONTRACTS.
Abstract: Data race occurs in a multithreaded program when several threads simultaneously access the same memory location and at least one of them has a write access. Data races are hardly reproducible and can damage global data structures; so, research in the area of automatic race detection methods has been carried out for more than 20 years. This article focuses on the issue of improving performance of dynamic race detection in Java programs without loss of precision. Synchronization contracts - partial specifications of multithreaded behavior - are introduced for solving this problem. Using contracts allows excluding parts of application's code that are not interesting from the race detection perspective (e. g., external libraries). The paper also covers advantages and restrictions of the approach, the contracts specification language, and some implementation details.
Keywords: multithreading; data race; dynamic analysis; automatic error detection
METHOD FOR EXTRACTING SINGLE-WORD TRANSLATION CORRESPONDENCES FROM PARALLEL TEXTS USING DISTRIBUTIONAL SEMANTICS MODELS.
- Yu. I. Morozova
- E. B. Kozerenko
- M. M. Sharnin
Abstract: The paper deals with problems of corpus research of linguistic units. The task of extracting translation correspondences from a parallel corpus is defined. An overview of existing approaches to this task is provided. The paper focuses on the approach to extracting translation correspondences based on distributional semantics models. The paper describes the theoretical model developed by the authors as well as its software implementation. A test parallel corpus of patent texts in French and English was compiled for the purpose of this research. The paper provides results of an experiment aimed at extracting single-word translation correspondences from the test parallel corpus.
Keywords: extracting translation correspondences; alignment; parallel texts; parallel corpus; distributional semantics; vector space model
ABOUT PROBLEMS OF IMPLEMENTATION OF SEMANTIC GEOINTEROPERABILITY IN SEMANTIC WEB.
- S. K. Dulin
- N. G. Dulina
- D. A. Nikishin
Abstract: The problem of semantic geointeroperability consists in supporting coordinated interaction of experts for solving tasks demanding georesources shared use, under condition of adequate understanding of semantics of the georesources. Support of semantic geointeroperability assumes that development of tools of coordinated understanding of geodata should be carried out on the basis of the comparative analysis of existing metaschemes of bases of geodata in view of multifactor interactions of users and semantics incorporated in space ontology and/or geothesauruses and qualifiers. The key task of semantic geointeroperabil- ity is creation of a uniform conceptual model of representation and coordinated understanding of geodata on the basis of integration of the space-distributed information. The paper is devoted to discussion of the concept of semantic interoperability, aspects of implementation of semantic geointeroperability of the geographical information, and standards of semantic geointeroperability.
Keywords: geodata; semantic geointeroperability; ontology; Semantic Web
ANALYTICAL ASPECTS OF MULTIAGENT DISTRIBUTED CONTROL SYSTEMS.
Abstract: TThe paper considers the issues of application of the multiagent approach for creation of integrated systems of situational management, taking into account their specific features: complexity of environment interaction, independence, autonomy, and decentralization. The paper studies organization of multiagent control systems and agent framework, implements a modified management cycle OODA (observation{orientation{decision{action) to create a system of distributed situational centers. The methods of data analysis used at all stages of themanagement cyclewere systematized.The structure of a collaborative environment for distributed control system was studied. The system exchanges not only event information, but also data analytics to support decision making while realizing the objectives of the control system. The ways of formalizing the analytical data were developed with regard to all components of the control area of the distributed control system, including targeted environment, controlled objects, controlled resources, and uncontrolled factors.
Keywords: multiagent system; situational center; distributed control system; analytical data
THE TECHNOLOGY OF INTERNET SEARCH AS A PART OF BIOGRAPHIC INVESTIGATION.
- I. M. Adamovich
- O. I. Volkov
Abstract: The article examines the role and specificity of Internet search as a part of biographic investigation. The reasons of possible problems occuring during search are explored. With this purpose, the types of user's biographical information necessities were analyzed. The specific biographical information necessities were found and the type of search associated with them | \indirect search" was described. The requirements for Internet search technology which supports and simplifies indirect search are formulated. An example of such technology which does not use any special software and is based on a set of rules is described and analyzed. The requirements for special search support software were formulated as the result of analysis of faults of this technology. The technology of effective biographical search based on such software, free of analyzed faults and supporting storage and access to search results was described. This technology is meant for a wide range of users which are not professional historians or biographers. This is important today because of increasing public interest in family history.
Keywords: biographic investigation; technology of Internet search
COMMUNICATORY PROVISIONS FOR MONITORING OF NATIONAL SECURITY FROM THE REGIONAL POINT OF VIEW.
- G. V. Lukyanov
- D. A. Nikishin
- G. F. Verevkin
Abstract: Monitoring and estimating of national security of the Russian Federation (NSRF) assumes processing of several tens of various parameters envisioned for this purpose. A priori it is supposed that these parameters are not subject to regional specificity or this influence is insignificant. The analysis carried out by the authors has shown that a significant part of parameters is injected from the regional components and can be received only from regional sources. Thus, these parameters vary seriously from region to region and their distinctions have not only quantitative, but also qualitative character. Essential distinctions in natural, climatic, and geological conditions are the major factors of regional specificity though some experts also specify unequal "starting" capabilities of various regions. One more feature of general statistics is reflected in its quite formal character, sometimes rather indirectly reflecting the essence of the phenomenon which the appropriate indices estimate. Only a profound study of the situation allows to straighten out the reason of such divergences. These circumstances dictate the necessity of working out specialized techniques considering regional specificity while calculating basic (federal) parameters of NSRF. One of the solutions could be applying the regional correction coefficient which would allow to correct a number of statistics that are the most significant from the point of view of NSRF in order to achieve better objectivity.
Keywords: communicatory provisions; monitoring; regional differentiation; regional features; national security; administrative and territorial structure; federation subject
METHODS OF CLOUD SERVICE INTEGRATION BY THE EXAMPLE OF HEALTHCARE.
Abstract: The article gives a review of traditional approaches to integration of many independent applications into a unified information system (IS), their merits and demerits. The approaches to integration of applications in the context of the unified IS and requirements for solution of this task based on constructing a special integration layer of an IS are described. The solution is built on the principles of the service-oriented architecture, allows to use modern integration products of different companies, and offers additional architectural tools of ranging. These tools give an opportunity of optimum loading of software and hardware tools of the integration layer, starting with small IS like enterprises and up to scaled geographically distributed IS. Particularly, the article shows possibility and expediency of using the integration layer mentioned above for implementing the unified state IS in the field of healthcare for serving the task of integration of its components into the unified IS.
Keywords: information systems; service-oriented architecture; web services; integration of components of IS; intermediate layer; unified state information system in the field of healthcare
APPLICATION OF WEB RESOURCES OF THE SINF INFORMATICS KNOWLEDGE SYSTEM IN EDUCATIONAL PROCESS.
Abstract: The article analyzes the experience of using web resources of the distributed hypermedia system of informatics knowledge (SINF) for teaching students. SINF consists of the journal "INFORMATICS: S-modeling" and the informatics encyclopedia "INFOPEDIYA." The journal publishes monographs, articles, tutorials, reviews, comments, and various notes. The encyclopedia provides definitions of informatics concepts and gives necessary explanations. The feature of SINF is continuous expansion of knowledge system by increasing the amount of available scientific and educational resources. It is shown that the SINF web resources, which are used in educational process, contribute to enhancement of its quality. The quality of teaching is also improved by combination of classroom and remote studies (video lectures, Skype sessions with lecturer).
Keywords: web resources; informatics knowledge system; journal "INFORMATICS: S-modeling," Encyclopedia of informatics "INFOPEDIYA," educational process
|
