We invite you to participate in The Ivannikov ISP RAS Open Conference, which will take place on November 22-23, 2018 in the main building of the Russian Academy of Sciences, Moscow, Russia: Moscow, Leninskiy prospekt, 32A.
Within this conference sections for the IT industries directions are held, in which ISP RAS has accumulated long-term experience (cyber security, data management, data analysis and other), carrying out fundamental researches, developments of innovative technologies and also realization of specific projects in introduction of the developed technologies in the industry together with industrial partners.
The conference is held with the support of IEEE and IEEE Computer Society, the Russian Academy of Sciences (RAS), Advanced Research Foundation (ARF), the Federal Service for Technical and Export Control (FSTEC Russia) and partners of ISP RAS (Samsung, Hewlett Packard Enterprise, DELL, NVIDIA and others).
To present your research at 2018 Ivannikov ISP RAS Open Conference please submit a paper, that wasn't published and not submitted for other conferences.
We solicit original and unpublished articles of current research, results or work in progress. We invite both researchers and practitioners to submit early research work or general discussions. All accepted papers will be published in the Proceedings of Ivannikov ISP RAS Open Conference.
Submission Guidelines is here.
To participate in the sections of Ivannikov ISP RAS Conerence please register.
Fees: free of charge.
The Ivannikov ISP RAS Open Conference is the annual event under which the Institute for system programming of the Russian Academy of Sciences (ISP RAS) arranges a number of conferences devoted to sectors of IT-industry in which it possesses a many years’ experience of fundamental researches conduction, innovative technologies development and implementation, together with partners, of certain projects on introduction the developed technologies in the industry.
The goal of this event is to provide support and development for the innovations generation and high-skilled system programming workers reproduction ecosystem, created in the ISP RAS as well as to create prerequisites for improving the level of using newest information technologies in activity of educational, scientific and research organizations and their further introduction in Russian industry. During The Ivannikov ISP RAS Open Conference sections reports on fundamental and applied researches as well as on issues related to new technologies introduction are made.
Acknowledged experts from national and foreign scientific and educational organizations, representatives of leading IT-companies and researchers whose studies have been professionally reviewed are invited to make reports.
The event is being held by joint initiative of the ISP RAS and is devoted to methods of optimization in compiler and code generation, static and dynamic program analysis, etc. Issues, examined under this event include, but are not limited to the following topics:
The event is intended for researchers specializing in data management and data analysis. We invite researchers working on methods and systems for storing and analyzing large volumes of text, graph, multimedia, scientific and other types of data. Issues addressed by the conference include, but are not limited to the following topics:
The event is being held by joint initiative of the ISP RAS and is devoted to technologies of blockchain and distributed ledgers. Issues to be considered during this event include, but are not limited to the following topics:
The event is intended for practical experience exchange between researchers, who use open applied packages for continuum mechanics problems solving. Issues being studied under this conference include but are not limited to the following subjects:
1 System Programming Laboratory at Russian-Armenian University, 2 ISP RAS
In this work, we tackle the problem of Armenian named entity recognition, providing silver- and gold-standard datasets as well as establishing baseline results on popular models. We present a 163000-token named entity corpus automatically generated and annotated from Wikipedia, and another 53400-token corpus of news sentences with manual annotation of people, organization and location named entities. The corpora were used to train and evaluate several popular named entity recognition models. Alongside the datasets, we release 50-, 100-, 200-, 300- dimensional GloVe word embeddings trained on a collection of Armenian texts from Wikipedia, news, blogs, and encyclopedia. Index Terms: machine learning, deep learning, natural language processing, named entity recognition, word embeddings
1 Lomonosov Moscow State University, 2 ISP RAS
1 ISP RAS
1 Kazan Federal University
This paper presents our experimental work on neural network models for entity-level adverse drug reaction (ADR) classification. Aspect-level sentiment classification, which aims to determine the sentimental class of a specific aspect conveyed in user opinions, have been actively studied for more than 10 years. In the past few years, several neural network models have been proposed to address this problem. While these models have a lot in common, there are some architecture components that distinguish them from each other. We investigate the applicability of neural network models for ADR classification. We conduct extensive experiments on various pharmacovigilance text sources including biomedical literature, clinical narratives, and social media and compare the performance of five state-of-the-art models as well as a feature-rich SVM in terms of the accuracy of ADR classification.
1 CICESE Research Center, 2 Tsinghua University, 3 ISP RAS, 4 UMET
Lightweight virtualization technology has emerged as an alternative to traditional hypervisor-based virtualization. Containers based on an operating system level virtualization have shown superior performance and more flexibility than virtual machines. Both factors encourage their fast adoption and wide use in cloud environments. Container technology guarantees efficient interaction with hardware through system calls and isolation of tasks execution. The portability and ease of deployment enable containers to execute a variety of jobs. Several techniques have been used to address the problem of job allocation on containers. In this paper, a set of allocation strategies is proposed to distribute the workload to multiple OS containers that are running on several servers. We address two objective scheduling problem: minimizing energy and a number of SLA violations. We propose seventy-two strategies and analyze their performance by comprehensive simulation. We demonstrate that our proposed techniques outperform previous works in terms of both the energy savings and QoS.
Aspect extraction from user reviews is one of the sources to make dialog systems, which are on the rise now. A typical user of a conversation system has no time to check the spelling or grammar in his or her utterances. Due to that user utterances contain typos and spelling errors, so the noise robustness should be considered as a significant feature of an aspect extraction model. We analyze noise-robustness of state-of-the-art Attention-Based Aspect Extraction technique and propose the extensions for this model, which lead to more robust behaviour in presence of typos. Experimental results demonstrate how suitable each of the complements to the model that uses the data containing typos. Index Terms: aspect extraction, noise robustness, neural networks.
1 Federal State Budgetary Educational Institution of Higher Education "Orenburg State University"
The problem of econometric models specifications, as well as the problem of analyzing mixed frequency data, arises in many issues related to the modeling of processes and real-world phenomena, in particular - the modeling of yields. The main problem is that despite the variety of ready- model specifications, it is necessary to rework factors of influence in order to apply them to a specific region, therefore, each time it’s necessary to re-verify and refine the model specification. In addition, it is often difficult to make a choice between models on panel data and an aggregate set of time series analysis models. The research methodology proposed allows us to determine the specification of models faster and more qualitatively; the developed software allows conducting research on various sets of initial data, obtaining qualitatively high results with an automatically selected specification of models. In the framework of this paper, we considered the singular spectral analysis algorithm as an indicator of seasonal components and the trend in time series of yield regressors, as well as we considered models on mixed-frequency data using MIDAS models and construction of models that take into account panel structure of the original data.
1 Ain Shams University
Combinatorial optimization is the search for an optimal configuration of a set of variables to accomplish certain goals. One of the well-known combinatorial optimization problems is the timetabling problem, with a lot of research conducted in the past few decades to investigate a variety of methodologies to solve it. One of the blossoming recent methodologies is hyper-heuristics, which attempts to automate the algorithm design process so that it would be able to work with different sets of problem domains. This paper focuses on the university course timetabling problem (UCTP) as the case of study, and proposes the use of a competitive iterated local search approach strengthened with an add-delete hyper-heuristic. The hyper-heuristic utilizes an adaptive heuristic generation mechanism through a variable-sized list of add and delete operations. The algorithm was enhanced with the use of a novel approach to construct a good feasible initial solution and strengthened with a diversifying mechanism to allow more exploration over large search spaces to find a global rather than local near optimal solution. The proposed work was tested with the ITC2007 benchmark datasets, and experiments show promising results and give better average performance when compared to recent approaches in the literature that work on similar timetabling problems.
1 Utrecht University
1 Bundeswehr University Munich
In this work numerical simulations were performed based on the experimental work of Wu el al. , who mentions fuel condensation effects during the high-pressure injection of ethylene into nitrogen at rest. In order to investigate the phase separation phenomenon numerically a hybrid pressure based solver was coupled with a consistent real-gas thermodynamics implementation and a phase separation model using the opensource framework of OpenFOAM. Three different injection temperatures were simulated confirming the experimental findings of Wu el al.  and illustrating, that lower injection temperatures lead to more significant fuel condensation effects. The occurring phase separation takes place due to the expansion of the fluid and the mixing process with the surrounding nitrogen. Index Terms: High-pressure injection, real-gas thermodynamics, phase separation, fuel condensation, hybrid pressure-based solver.
1 ISP RAS
1 ISP RAS
1 Institute of Computational Mathematics and Mathematical Geophysics SB RAS, 2 Institute of Astronomy RAS
The paper presents the results of mathematical modeling of the three main stages of the explosion of supernovae of type Ia (SNIa): merging of white dwarfs, collapse and noncentral explosion. In the fusion problem shown the general fall mechanism of the satellite star onto the central star. At the modeling of the collapse process, the behavior of various types of energies is investigated. Also shown is the qualitative dynamics of a noncentral of a star explosion, whose source is the characteristic value of the energy released by the reaction C + C. The simulation was carried out using the original HydroBox3D code, in which the concept of adaptive nested mesh is implemented. Index Terms: Adaptive nested mesh, Computational Hydrodynamics, Computation Astrophysics, Numerical simulation, SNIa.
1 Marchuk Institute of Numerical Mathematics of the RAS
We present the INMOST library - a C++ based software platform for parallel mathematical modeling. Platform includes data structures and algorithms for distributed mesh and data representation; automatic differentiation for (block) assembly of residual vector, jacobian matrix and hessian tensor; several parallel linear and nonlinear solvers; preliminary support for complex coupled problems. The platform was extensively used in development of nonlinear finite volume methods for scalar diffusion and advection-diffusion equations. INMOST is used in software with complex physical models such as geomechanics, black oil recovery, hemodynamics, nuclear waste disposal.
1 NSI RAS, 2 Queen Mary University of London
1 Keldysh Institute of Applied Mathematics
The article is devoted to a comparative evaluation of the accuracy of solvers of the OpenFOAM software package. As the test problem, an axisymmetric problem of flow around a cone by an inviscid flow of a compressible gas at zero angle of attack is used. Simulation results for different solvers are compared with the known numerical solution of this classical problem. When solving the test problem for each solver, the angle of the cone and external flow rate are varied with a certain step. The comparison is carried out in a fairly wide range of these determining parameters. Special attention is paid to the QGDFoam solver, which has the property of suppressing undesirable oscillations in the vicinity of the cone vertex while maintaining sufficient accuracy.
1 ISP RAS, 2 Samsung Electronics, Software Center
In the previous publication , we described a method for minification of a system composed of dynamic modules under the "closed-world" assumption. Although the presented approach is sound in theory, there are complications with reliably using it in practice. Such complications arise from the peculiarities of existing real-world linkers (symbol elimination relies on proper and consistent implementation of the plugin infrastructure and the --gc-sections switch). The above approach is also imprecise in that it keeps more symbols than strictly necessary due to not discriminating between build time only dependencies (from configure tests) and regular ones. This design limitation as well as the reliability issues are addressed in the new approach presented in this paper.
1 ISP RAS
Many buffer overrun errors in C programs are caused by erroneous string manipulations. These can lead to denial of service, incorrect computations or even exploitable vulnerabilities. One approach to eliminate such defects in the course of program development is static analysis. Existing static analysis methods for analyzing strings either produce many false positives, miss too many errors, scale poorly, or are implemented as a part of a proprietary software. To cover a significant amount of the real program defects it is necessary to detect errors that could happen only on a particular program path and cannot be defined by a single erroneous point. Also, it is essential to find misusage of library functions and user functions. The aim of this study is to develop a detection algorithm that will cover such cases, will produce at most 40% false warnings, will be applicable to any C programs without any additional restrictions, and will scale up to millions of lines of code. We have extended our approach of symbolic execution with state merging to support string manipulations. Also we have developed a string overflow detector based on our buffer overflow approach with integer indices. The new algorithm was implemented in the Svace static analyzer. As a result, the coverage of the buffer-overflow related testcases from the Juliet test suite has increased from 15.4% to 41.5% with zero false positives. Also we have compared our Juliet results to those of the Infer static analyzer. The basic Svace version (without string support) is on par with Infer except for the flow variant of complex loops, whereas string-related buffer overflows are not detected by Infer.
1 ISP RAS
Existing standards for airborne-embedded software systems impose a number of requirements applicable to the software development cycle of hard real-time operating systems found in modern aircraft. The measures taken are meant to reduce the risks of undesired consequences, but have strongly varying costs. Dynamic instrumentation and static analysis are common practices used to automatically find software defects, from strictly non-conforming code constructions to memory corruptions or invalid control flow. LLVM analyser and sanitizer infrastructure, while regularly applied to general-purpose software, originally was not thought to be introduced to heavily restricted environments. In this paper we discuss the specifics of airborne systems with regards to dynamic instrumentation and provide practical considerations to be taken into account for the effective use of general-purpose instrumentation tools. We bring a complete LLVM stack support to JetOS, a prospective onboard real-time operating system currently being developed at ISP RAS in collaboration with GosNIIAS. As an example, we port AddressSanitizer, MemorySanitizer, and UndefinedBehaviorSanitizer and provide the details against the caveats on all relevant sides: a sanitizer, a compiler, and an operating system. In addition we suggest uninvolved optimisations and enhancements to the runtimes to maximise the effects of the tools. Index Terms: dynamic instrumentation, real-time operating systems, ARINC 653, IMA, LLVM.
1 NRU HSE, 2 ISP RAS
Abstract—This paper describes the development of an auto-active verification technique in the Frama-C framework. We outline the lemma functions method and present the corresponding ACSL extension, its implementation in Frama-C, and evaluation on a set of string-manipulating functions from the Linux kernel. We illustrate the benefits our approach can bring concerning the effort required to prove lemmas, compared to the approach based on interactive provers such as Coq. Current limitations of the method and its implementation are discussed.
1 ISP RAS
An analysis of the emergence of computer science, the first computer and computer science is given. The issues of computerization and industrialization of computer systems support and technologies for the development of applied systems for various areas of industry (aviation, space, technical, etc.) and the training of scientific personnel until 1992 are considered. After 1992, the era of informatization and commercialization of domestic and foreign fields of activity of the world community began. Computer systems for automation and intellectualization of various types of industry, medicine, education, etc. are given. New types of strategies for the development of the economy and education in our country are defined. New, more progressive methods of computerization of system software and programming paradigms are considered. The prospects for the development of computer science, informatic and computer technology are defined.
1 Academy of the Federal Guard Service
1 Innopolis University, 2 A.P. Ershov Institute of Informatics Systems
Svace is an essential tool of the secure software development life cycle, the main static analyzer that is used in Samsung Corp. It detects more than 50 critical error types. Svace supports C, C++, C#, and Java. Svace is registered in the National Software Unified Register, which is kept by the Ministry of Digital Development.
BINSIDE is a tool for detecting flaws in a program using static analysis of executable code.
Anxiety is a framework for detecting errors and potentially dangerous cases in the process of development, acceptance testing and operating the software.
Binary code analysis platform based on QEMU emulator. The ISP RAS platform for program analysis is built on the basis of an open QEMU emulator, which is used when crossplatform development is needed.
ISP Obfuscator is a set of technologies to prevent mass exploitation of vulnerabilities resulting from errors or bookmarks. If the hacker is able to attack one of the devices with the common software, the rest will remain protected by changes made to the code.
Protosphere is a system of deep packet inspection (DPI). It is the part of intrusions and information leaks protection. Detects inconsistencies between protocol specification and specific implementation. Allows you to quickly add support for new (including closed) protocols due to the flexibility of the internal representation.
Klever is a static verification system that uses advanced tools to thoroughly check the security, reliability and performance of software systems developed in the GNU C language. In particular, it is used to verify the real-time OS.
MicroTESK is a reconfigurable and expandable test program generation environment for functional microprocessors verification. It allows automatically constructing test program generators for target microprocessor architectures based on their formal specifications. MicroTESK is applicable for a wide range of architectures (RISC, CISC, VLIW, DSP).
Retrascope is a tool for reverse engineering and functional verification of digital equipment descriptions. It provides automated tools for extracting and analyzing formal source code models. The tool supports synthesized subsets of Verilog and VHDL languages.
AstraVer Toolset is a system for deductive verification of key components. It allows developing and verifying security policy models, as well as providing evidence of the correctness of key components in C language. The necessary tool to achieve the goals of the ADV_SPM and ADV_FSP trust families defined in GOST R ISO / IEC 15408-3-2013.
MASIW is a set of tools for developing software and hardware packages for mission-critical systems in the field of aviation, medicine, and others. It is created for design engineers of aircraft on-board equipment developed using integrated modular avionics (IMA). It quickly adapts to other subject areas.
Constructivity 4D is a technology for creating ambitious software systems and services operating with large arrays of spatio-temporal data and dynamic scenes. It is able to conduct a visual analysis of millions of objects with different geometric representations and individual dynamic behavior. The technology is implemented in the Synchro system, designed for 4D-modeling of large industrial sites.
Texterra is a scalable platform for extracting semantics from text. It is the basic set of technologies for creating multifunctional applications. It analyzes texts using concept identification. It is included in the Unified Register of Russian software.
Talisman – big data processing solution for social and commercial information retrieval. It recognizes patterns in relationships by analyzing large graphs from hundreds of millions of nodes.
Lingvodoc is a system intended for collaborative multi-user documentation of endangered languages, creating multi-layered dictionaries and performing scientific work with the received sound and text data. Joint project with the Institute of Linguistics of the Russian Academy of Sciences and Tomsk State University. Under development since 2012. Project website – lingvodoc.ispras.ru.
SciNoon is a system for researcher’s inquiry of scientific articles. Combines a number of unique features to optimize the process of searching and analyzing the results. It particularly allows you to work in a team and keep a history of user actions. Can work with big data.
Complex of solutions for creating service-oriented data processing center provides the ability to store data and perform complex, resource-intensive calculations using both containers and virtual machines. It is particularly intended for the deployment of cloud environments.
1 Pax Datatech, Seoul, South Korea
1 Tallinn University of Technology
The application and use of smart contracts in organizations require a holistic overview. This overview helps to understand the current adoption of this technology and also deduces factors that are inhibiting its use in the modern organization. This study provides a systematic review of previous studies comprising of frameworks, methods, working prototypes and simulations that demonstrate the application of smart contracts in organizations. Understanding the current state and usage of smart-contract technology in an organization is a focal point of this paper. Much progress occurs in developing technologies that support smart contracts, while little understanding exists pertaining to their usage in organizations. In this study, we identify properties of smart-contract applications in different domains of modern organizations. We further analyze and categorize challenges and problems mitigating the adoption of smart-contract applications.
1 KCD LLC
With the development of Internet technologies, the number of threats and attacks directed at networks and systems is increasing. Attackers invent new ways of attack or improve old ones. One of the most common serious threats is "Phishing", in which cybercriminals attempt to steal user credentials using fake emails or websites or both. Blockchain projects increasingly becomes the target of attacks by intruders due to high investments and gaps in national legislation. After a series of attacks on blockchain projects around the world, the issue of cybersecurity of blockchain became particularly relevant. The article classifies the main types and schemes of phishing attacks on the blockchain, suggests methods of protection against them, examines the development of blockchain protection against phishing attacks.
1 Skobeltsyn Institute of Nuclear Physics, Lomonosov Moscow State University, 2 Yaroslav-the-Wise Novgorod State University
The paper suggests a new approach based on blockchain technologies and smart contracts to creation of a distributed system for managing provenance metadata, as well as access rights to data in distributed storages, which is faulttolerant, safe and secure from the point of view of preservation of metadata records from accidental or intentional distortions. The implementation of the proposed approach is based on the permissioned blockchains and on the Hyperledger Fabric blockchain platform in conjunction with Hyperledger Composer. Index Terms: distributed storage, provenance metadata, blockchain, access rights, Hyperledger.
1 Dexpa LLC
In our paper, we are researching an application of the Tendermint protocol for designing a loyalty program. The proposed approach allows operating by pseudonyms in a public network and unique non-fungible tokens. These tokens could be tagged and sent to "issuer" of the various loyalty programs. In that case, we can ensure that pieces of data are revealed selectively and clients could be safely de-anonimysed at a certain level and via certain service or organization exclusively.
1 Pax Datatech
1 Marchuk Institute of Numerical Mathematics RAS, 2 Moscow Institute of Physics and Technology, 3 Sechenov University
1 The Nuclear Safety Institute (IBRAE)
State-of-the-art models of dispersion of contamination in the urban environment and industrial areas employ a CFD approach in order to calculate turbulent characteristics of flow around buildings with complex geometry. The main area of application of these models is to give assistance in building licensing of potentially hazardous facilities and assessment of meteorological conditions in the urban environment. The usage of the most popular commercial CFD software with regard to modelling flows in the urban environment is significantly limited by the requirement for computational mesh refinement near the surface of the building in order to adequately resolve the characteristic scales in the viscous and intermediate sublayer. On the other hand, models based on traditional gaussian approach cannot take into account the complex aerodynamic effects in order to calculate turbulent characteristics of flow around buildings with complex geometry, including the subtleties when dealing with atmospheric emissions of gas-aerosol substances. Therefore, the authors developed a robust, highly specialized CFD-RANS model and a calculation code for modelling the atmospheric dispersion of contamination under conditions of a complex three-dimensional geometry that do not require mesh refinement. The authors carried out verification of this model using extensive database obtained both in the course of field experiments as well as of wind tunnel experiments. The verification results showed that the developed model satisfies the acceptance criteria for the quality of modelling along with foreign general-purpose codes and highly specialized codes.
1 BMSTU, 2 ISP RAS
1 Shirshov Institute of Oceanology, 2 Dorodnicyn Computing Center FRC CSC RAS
Estimates of the meridional mass flow in the Antarctic bottom water, calculated from the coupled Earth-Earth-atmosphere model, MPI-ESM and using original methods of observational data assimilation are presented. For the assimilation data were acquired of latitudinal CTD temperature and salinity profiles in 1991-1995 taken from the World Ocean Circulation Experiment WOCE. Estimates of the velocities of the Antarctic bottom water currents as purely model (control) and with the assimilation of observational data are given. The assimilation method previously called as the generalized Kalman Filter (GKF), was used, which coincides in a particular case with the classical Kalman method (EnKF). It is shown that the model calculations are close to estimates of the geostrophic flow transfer based on the dynamic method.
1 Institute of continuous media mechanics UB RAS
The work deals with numerical simulations of multiphase flow and displacement front in porous media. In practice, the displaced fluid often forms compact inclusions surrounded by the displacing fluid. In this case, we should consider displacement fronts on both forward and back sides of the inclusion. The drop is initially spherical and the interface thickness between liquids is supposed to be small (much less than the drop radius). The calculations are carried out using the Darcy model by Level set method with adaptive mesh refinement algorithm that dynamically refines computational mesh near interface. The open source software PARAMESH is used to operate with the adaptive mesh. Numerical modelling of the droplet dynamics subjected to gravity has shown that the instability develops at the forefront of moving droplets regardless of the ratio of fluids viscosities. Under modulated pressure gradient small-scale perturbations of interface are suppressed and in the case of modulation with large enough intensity drop becomes stable. Index Terms: adaptive mesh refinement, porous media, paramesh, aztec.
1 ILIT RAS - Branch of the FSRC "Crystallography and Photonics" of RAS
The developed hydrodynamic model of laser cladding (LC) considering catchment efficiency is verified with experimental data for steel powder LC in a wide range of process parameters. For direct numerical simulation of the process, open computational fluid dynamics package OpenFoam is used. PyFoam is used to automate numerical research. The key features of the OpenFoam code are presented. The comparison of the main calculated output parameters (depth of the penetration, track width and height) with experimental data is held.
1 KIAM RAS
The methodology and the results of numerical simulation of flows in the scramjet combustor using the open source CFD-toolbox OpenFOAM are provided. The combustor chamber is supplied with gas phase hydrocarbon fuel. The dynamics of combustion processes is studied depending on oxidizer-to-fuel ratio. The system of the Reynolds-averaged Navier-Stokes equations with additional terms and equations taking into account the effects of turbulence and combustion is used as a mathematical model.
1 State Marine Technical University
The paper shows results of computational fluid dynamics possibilities for prediction motions of ship with moonpool and vertical water motions in moonpool in regular head waves with zero ship speed. Moonpool is well which is used in different types of ships such as cable laying, drill and FPSO, survey, research and so on. This well is used for launching and lifting of different devices, divers, rescue bells, cables and risers which are protected of outboard wind and waves. Results of numerical simulation in OpenFOAM source of heave and pitch motions of DTMB 5415 model in regular head waves with and without ship speed show good agreement with experimental data. Experiment was organized with series 60 model which was equipped different moonpool shapes modules in regular head waves with zero ship speed for determine heave and pitch RAOs and vertical water motions in moonpool. Results don’t show any influence of moonpool for heave and pitch ship motions. These data are necessary for numerical simulation verification. Results of numerical simulation of experimental research show good agreement that means good efficiency of computational fluid dynamics in heave and pitch motions and vertical water motions in moonpool calculation of ship with moonpool. Numerical simulation should be advised for calculations during ship design process.
1 ISP RAS, 2 BMSTU, 3 Keldysh Institute of Applied Mathematics
The paper is concerned about the different ways to speed up computations in discontinuous Galerkin solution of two-dimensional gas dynamic problems on unstructured meshes. An optimal choice of the fastest tools for numerical scheme construction and parallelization of computations make numerical simulation much faster. An influence of choice of numerical algorithms and parallelization technology to speed and accuracy of computations has been investigated on the C++ code which implements two ways of temporal discretization (Runge — Kutta and Adams approaches), the set of numerical fluxes (LF, HLL, HLLC), some methods of solution monotonization (limiters based on the WENO approach and the local maximum principle). Two parallel version of code based on OpenMP and MPI techniques are compared.
1 Keldysh Institute of Applied Mathematics
Supercomputer simulations are of fundamental importance for understanding the physics of nonlinear processes in high temperature pulse plasma. The development of predictive codes is an urgent problem in computational plasma physics, as well as many other fields of science. Radiative magneto-hydrodynamics 3D code MARPLE is a full-scale multiphysics research code using the state-of-the-art physics, mathematics, and numerics as well as the up-to-date high performance computing functionality. Scalability study demonstrated that the code can fit existing petaFLOPS supercomputers as well as next-generation exaFLOPS ones. The code is currently used for multiphysics simulations, specifically for high energy density plasma in pulsed-power facilities. Compression of a wire array by a high-current discharge is a valuable tool for fundamental study of matter in extreme states. Different configurations of wire arrays were investigated numerically. A series of high resolution computations helped to create a very compact spherical bright radiation source using dedicated design of the electrodes, the wire array, and the mass distribution along the wires.
1 ITMO University, 2 KIAM RAS
The concept and principles of the organization of Neuro-Fuzzy traffic control of transport systems on the basis of intelligence technologies and high-performance means of computing is discussed. The developed information model provides the operative control of the current situations caused by transport streams in difficult dynamic environments. The special attention addresses on the decision of a problem of planning of operations and development of operating decisions at realization of a competition principle in the conditions of uncertainty and incompleteness of the initial information.
On the basis of the RDF-storage, a software tool was developed for creating a knowledge base containing information about the CFD-calculations performed. The software is a set of scripts written in bash and python, which are published under the GNU GPL3 license. The tool is designed to support the user when making research studies that do not have a strict, pre-defined design of experiment or problem solving algorithm. To formalize the description of the calculation stored in the knowledge base, an ontology is created that serves as the information model for the calculation. As an auxiliary mechanism for carrying out an automated comparison of calculations with each other (a mechanism of "comparators" and "features") was developed and also described in the article. In addition to the systematized data storage, the complex provides the possibility of their automated and semi-automated analysis, including the presentation of a set of calculations in the form of an undirected graph, the construction of flat and spatial dependencies, the search for similar calculations, etc. The article gives examples of data processing results for the project on design of channel in the cylinder head of a piston engine.
1 Kalashnikov ISTU, 2 JSC "INNTS", 3 UdFRC UB RAS
Models (of different completeness) of gas-dynamic processes in direct acting safety valves are considered. All the models are based on the control volume method in 3D, axisymmetric or 1D formulation. In 3D simulation, the internal cavity of the valve is divided into two blocks. In each block, a structured difference mesh is constructed. The mesh in the first block is orthogonal. The proposed approach to numerical simulation of non-stationary 3D gas dynamics enables to determine a spatial structure of flow in a safety valve and its quantitative characteristics (pressure, density, velocity, temperature). To determine the spatial structure and calculate the dynamics of quantitative characteristics of the valve, the first (comprehensive) level of simulation must be applied. The axisymmetric formulation of the gas-dynamic problem can be applied at the stage of disc lifting (valve actuation). Assumption of axial symmetry of the flow under the disc is also justified in simulating stationary characteristics of safety valve. It is a good practice to study wave processes in supply and discharge lines using simplified axisymmetric approach, which is comparable with 1D calculation in terms of computational burden.
1 Petrozavodsk State University, 2 Institute of Applied Mathematical Research KarRC RAS
A residential wood log stove design is one of the areas of industry where the numerical simulation of the combustion process can be used to reduce the cost of the design. However, experiments are not much more expensive, than simulation in this area. Furthermore, the initial step of simulation consists of developing or buying software and hardware products and setting up the model parameters that is very expensive therefore the simulation is seldom used in this area. To solve this problem it is necessary to develop the numerical solver with the open-source software product. In the present study we use the computational fluid dynamics (CFD) method for simulation the physical phenomena of the stove. The open-source software products OpenFOAM and BOINC were used for calculations. The optimization of the stove parameters such as efficiency and excess air factor was conducted. The results of the simulation are in good agreement with the experimental data. Thus, the results of this study can be used in the residential wood log stove design to reduce the cost. In the future work it is planned to add the pyrolysis model and more accurate simplified mechanism for model the wood combustion in order to increase accuracy of the simulation. Also we plan to consider complex geometry of the chimney and to preparation the BOINC-environment to automatic computing optimal parameters for a user-defined stove model.
1 Kazan Federal University
The paper is devoted to the study of the interaction of thin long beams with air in resonance modes of vibration. Numerical and experimental investigations of the the problem are carried out.
1 Novgorod State University
1 ISP RAS
Providing security for computer programs is one of the paramount tasks nowadays. Failures in operation of program software can lead to serious consequences and exploitation of vulnerabilities can inflict immense harm. Large corporations pay particular attention to the analysis of computer security incidents. Code-reuse attacks based on return-oriented programming are gaining more and more popularity each year and can bypass even modern operating system protections. Unlike common shellcode, where instructions are placed consequently in memory, ROP chain contains of several small instruction blocks (gadgets) and uses stack to chain them together, which makes analysis of ROP exploits more difficult. The main goal of this work is to simplify reverse engineering of ROP exploits. In this paper I propose the method for analysis of code-reuse attacks, which allows one to split chain into gadgets, restore the semantics of each particular gadget, and restore prototypes and parameters values of system calls and functions called during the execution of ROP chain. Parametrized types define gadget semantics. Each gadget type is defined by a postcondition (boolean predicate) that must always be true after executing the gadget. The proposed method was implemented as a program tool and tested on real ROP exploits found on the internet.
1 ISDCT SB RAS
The language FlexT (Flexible Types) is intended for specification of binary data formats. The language is declarative and designed to be well understood for human readers. Its main elements are the data type declarations, which look very much like the usual type declarations of the imperative programming languages, but are more flexible. While the primary purpose of the language FlexT development was to make the binary data understandable by displaying them according to the format specifications, recently we have implemented the code generator, which can produce data reading code in some imperative languages from the specifications. By now we have implemented the code generation for the most widely used FlexT data types, but some complex types like that used in specifications of machine instructions’ encoding are not supported yet. The current capabilities of the code generator are well characterized by that it have successfully produced a full-featured data reader code for the well-known for the GIS community Shape file format. In the article we first give a review of the features of the modern projects oriented to specification of binary file formats and consider their code generation capabilities. Then we briefly review the main features of the FlexT language. Finally we describe the main approaches that we use for the code generation.
1 Yaroslav-the-Wise Novgorod State University
The paper proposes the approach to file operations monitoring through capturing of the virtual disk accesses in the emulator. This method allows to obtain information about file operations in the OS-agnostic manner, but requires a separate implementation for each file system. An important problem for implementing this approach is the correct handling of changes in the file system. Operating systems that cache write requests, can perform operations in any order. We created a method for detecting read, write, create, delete and rename operations, and a module for QEMU, which monitors operations in the ext3 file system. Our method correctly processes the unspecified order of queries to the disk. The created module works in real time without the impact to the execution of the guest OS. The advantage of this method over others is that it does not interfere with the operation of the OS and does not depend on it. This allows it to be used on such exotic OSes that other monitoring tools for file operations do not work with. It is assumed that the QEMU module for file systems other than ext2 / 3 can be implemented using the methods described in this article.
1 Russian-Armenian University, 2 ISP RAS
In this paper we present a new approach for directed fuzzing. It enables us much faster generation of the input data for the target program’s specific instructions execution. Original fuzzing tools randomly generate or mutate input data to increase code coverage. This approach is not effective for analysis of special code regions. The basic idea behind of this paper is to instrument target program in a such way that interesting code fragments were executed as soon as possible. For that propose we detect all the paths in the program which are connecting program’s entry point to the considered instructions. Then we apply two type of instrumentations. In the first case we insert coverage collection instructions only in detected paths, which enables fuzzing tool to consider generated or mutated input data valuable if the distances between executed blocks and target points are reduced. In the second case we additionally insert 'exit(0)' instructions in those basic blocks from which target points are unreachable and their execution has no any influence. It allows repeatedly increase fuzzing speed.
1 ISP RAS
1 Mozhaiskiy Military Space Academy
The article presents the models and methods to implement the process of end-to-end quality control of complex software and hardware systems through the implementation of software-controlled process of development and verification of formal models of requirements and architecture of such systems. The analysis of scientific publications and normative-methodical base in the field of development and application in practice of model-oriented approach is given. It is established that least provided by model, algorithmic and software solutions are issues related to the development of a complete and correct set of requirements, as well as the formalization and verification of technical projects software and hardware systems. To solve the existing problems is proposed follow: develop of a unified language, modeling, information and software environment and use it for the development and verification of formal models of requirements and architecture of complex software and hardware systems, built on the basis of an optimal set of interconnected executable FUML diagrams, presented in the Alf language notation and verified in the environment of the virtual machine FUML using SMT and SAT solvers.
1 ISP RAS
This article describes the developed platform for static analysis of binary code. The platform is developed on the basis of interprocedural, flow-sensitive and context-sensitive analysis of the program. The machine-independent language REIL is used as an intermediate representation. In this representation basic data flow analyzes are developed and implemented - reaching definitions analysis, construction of DEF-USE and USE-DEF chains, analysis for deletion of dead code, value analysis, taint analysis, memory analysis and etc. The implemented approach for functions’ annotations allow data to be propagated between function calls, thereby making the context-sensitive analysis. The platform provides an API for using all implemented analyzes, which allows adding new analyzes as plugins.
1 ISP RAS
Nowadays software developers often face the following problem: there is a big amount of inputs which cause program to crash. In practice this amount of inputs is too large to analyze manually in a reasonable time. This article contains overview and analysis of existing methods for this problem. A new method is offered for analyzing crashes in order to select unique defects. The method is based on comparison of control flow graphs. For comparison of graphs a special metric is introduced: graphs are similar if the metric is not more than a threshold, which is a parameter. Information about the graphs is collected dynamically at runtime via instrumentation of the program’s binary code. The method is applicable to binary executables and does not need any debugging information. Developers, estimating their effort and time, can significantly decrease amount of crashes to analyze. The article also offers an algorithm for effective fixing of program bugs which cause crashes. The method has been implemented as a part of the ISP RAS fuzzer and tested on a set of programs for x86-64/Linux. The results of testing show that number of crashes for analysis can be decreased several times.
1 Positive Technologies
The article discusses ways to get the content of files which are modified during the processing in the well-known open source dynamic analysis environment Drakvuf. Drakvuf initially implemented file saving functionality based on the use of undocumented mechanisms for working with the system cache. The author of this article proposes a new approach to obtaining the content of files on Microsoft Windows family systems using Drakvuf. The proposed approach is based solely on the use of the public interface of the kernel by the hypervisor and provides portability between different versions of the operating system. In the conclusion of the article, the advantages and disadvantages of both approaches are presented, and directions for further work are proposed.
1 ISP RAS
There is still a gap between rapid development of new verification techniques and their practical application. One of major obstacles to performing sound formal verification of large GNU C programs is the necessity to prepare environment models. Researchers usually propose laborious ad-hoc solutions for environment modelling. Also, few software verification frameworks automate this step but they support a narrow class of software, e.g. device drivers or embedded systems. This paper proposes a method for automated compositional generation of environment models that supports adapting to project specifics and enables scalable software verification of various software. We evaluated the proposed method on device drivers and subsystems of the Linux kernel as well as on BusyBox applets.
1 Lomonosov Moscow State University, 2 Keldysh Institute of Applied Mathematics of the RAS, 3 ISP RAS
1 SPbPU, 2 The Ioffe Institute
1 ISP RAS, 2 JSC VNIIZHT
1 MIPT, 2 ISP RAS, 3 MAI
1 BMSTU, 2 ISP RAS
1 ISP RAS, 2 NRC Kurchatov Institute
Paper presents some results of the implementation of a quasi-hydrodynamic (QHD) approach as finite volume method (FVM) solver mulesQHDFoam on the basis of OpenFOAM. Application of the QHD numerical algorithm to the simulation of the attractor of internal gravity waves is considered. A comparison of FVM with spectral element method (SEM) implemented in Nek5000 is given. Convergence of the QHD FVM solver model to the SEM Nek5000 model is shown. The Big Data analysis method (Proper Orthogonal Decomposition) is used as a tool for comparing the calculation results between QHDFoam and Nek5000. Index Terms—Regularized gas dynamic equations, Quasi hydrodynamic equations, incompressible flows, Finite Volume Method, QHDFoam, mulesQHDFoam, MULES, FCT technique, attractor of internal gravity waves.
1 FSBEI HE «Kalashnikov Izhevsk State Technical University»
1 FSBEI HE «Kalashnikov Izhevsk State Technical University»
1 Marchuk Institute of Numerical Mathematics RAS
The list of keynote speakers is formed.
We solicit original and unpublished articles of current research, results or work in progress. We invite both researchers and practitioners to submit early research work or general discussions. All accepted papers will be published in the Proceedings of Ivannikov ISP RAS Open Conference.
The papers should be 3 to 7 pages long, formatted using IEEE Manuscript Templates for Conference Proceedings. The proceedings will be published online in IEEE Xplore digital library. Peer review is double-blind. Please remove any references to authors, such as names, affiliations or emails, from submitted manuscripts, as well as direct references to previous work. In the case of successful review and acceptance you will be requested to submit camera-redy text with necessary author details.
Articles submission due to September 19, 2018
September 5, 2018.
Submission is open at EasyChair.
Organizing and Program Committees of Ivannikov ISP RAS Open Conference look forward for cooperation with authors to ensure a process that is both fair in practice and perceived to be fair by everyone.
Submissions that violate these guidelines will be rejected.
Where to find us
The conference takes place at the Main building of the Russian Academy of Science. See the map for exact information.
To get to the venue place, you can take one of the city bus lines or railways Moscow Metro. All the services are comfortable to reach into the place.
Suitable Hotels and Restaurants are surrounded all the area. Here you will find our recommendations.