检索结果-内蒙古大学图书馆

Experiments with mpC: Efficient solving regular problems on heterogeneous networks of computers via irregularization

引用

5th International Symposium on Solving Irregularly Structured Problems in Parallel, IRREGULAR 1998

作者： Arapov, Dmitry Kalinov, Alexey Lastovetsky, Alexey Ledovskih, Ilya Institute for System Programming Russian Academy of Sciences 25 Bolshaya Kommunisticheskaya str. Moscow 109004 Russia

ISBN: (纸本)3540648097

mpC is a medium-level parallel language for programming heterogeneous networks of computers. It allows to write libraries of parallel routines adaptable to peculiarities of any particular executing multiprocessor system to ensure efficient running. The adaptable routines distribute data and computations in accordance with performances of participating processors. In this case even the problems traditionally considered regular, become irregular. Advantages of mpC for efficient solving of regular problems on heterogeneous networks of computers are demonstrated with an mpC routine implementing Cholesky factorization, with efficiency of the mpC routine being compared with ScaLAPACK one.

关键词： Heterogeneous networks

来源：评论

学校读者我要写书评

暂无评论

MPI-Based PFEM-2 Method Solver for Convection-Dominated CFD Problems 16th

MPI-Based PFEM-2 Method Solver for Convection-Dominated CFD...

引用

16th International Conference on Parallel Computational Technologies, PCT 2022

作者： Popov, Andrey Marchevsky, Ilia Bauman Moscow State Technical University Moscow Russia Ivannikov Institute for System Programming of the RAS Moscow Russia

ISBN: (纸本)9783031116223

The description of the parallel solver of the Particle Finite Element Method, 2nd generation (PFEM-2), is given. Strategies for the parallelization of both mesh-related and particle-related substeps are outlined. The software implementation is based on the open-source FEM code ***. The parallel solution of incompressible Navier–Stokes equations with the excluded convective term is performed with the out-of-box tools of *** using additional libraries such as Trilinos and p4est. Several MPI-based subroutines are developed for particle transport processing, as well as for particle/mesh field projection operations. The presented PFEM-2 solver allows for simulating convection-dominated flows with a high CFL number on relatively coarse meshes. The submesh resolution of the velocity field is maintained by particles. The results of simulation and speed-up of computations for test problems using multi-core/multi-processor systems are shown. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Finite element method

来源：评论

学校读者我要写书评

暂无评论

Validation of the developed open source library for far-field noise prediction 27

Validation of the developed open source library for far-fiel...

引用

27th International Congress on Sound and Vibration, ICSV 2021

作者： Epikhin, Andrey Ivannikov Institute for System Programming RAS Moscow Russia Bauman Moscow State Technical University Moscow Russia

ISBN: (纸本)9788378807995

The paper is focused on the validation of the developed open-source library libAcoustics for the far-field noise prediction by means of integral acoustic analogy. The library uses OpenFOAM for the near-field flow calculation and is available online at GitHub. Two benchmark cases with different formulations for computing the Ffowcs Williams and Hawkings equation are considered. In the first one, the spectrum of the acoustic noise generated by a rod-airfoil configuration is investigated. The near-field is calculated using large eddy simulation with a low-dissipation scheme, the far-field noise is predicted by a formulation based on the Garrick triangle. In the second case, acoustic perturbations generated by the compressible jet at a low Reynolds number are analyzed. Acoustic noise calculations are performed for the jet with Mach 0.9 and Reynolds number 3600 using Farassat 1A formulation. The obtained time-averaged and pulsating flow parameters and far-field noise have been compared with the available experimental data. The results of numerical simulations show a good agreement with the reference data. Thus, the developed library exhibits the potential for solving a wide range of computational aeroacoustics problems. © "Advances in Acoustics, Noise and Vibration - 2021" Proceedings of the 27th International Congress on Sound and Vibration, ICSV 2021. All rights reserved.

关键词： Acoustic noise

来源：评论

学校读者我要写书评

暂无评论

Model-based testing of safety critical real-Time control logic software 7

Model-based testing of safety critical real-Time control log...

引用

7th Workshop on Model-Based Testing, MBT 2012

作者： Gerlits, Yevgeny Khoroshilov, Alexey Institute for System Programming of the Russian Academy of Sciences lexander Solzhenitsyn st. 25 Moscow1009004 Russia

The paper presents the experience of the authors in model based testing of safety critical real-Time control logic software. It describes specifics of the corresponding industrial settings and discusses technical details of usage of UniTESK model based testing technology in these settings. Finally, we discuss possible future directions of safety critical software development processes and a place of model based testing techniques in it. © Y. Gerlits & A. Khoroshilov.

关键词： Model checking

来源：评论

学校读者我要写书评

暂无评论

Analysis of wake around DARPA Suboff including self-propulsion with a seven-blade propeller in positive pitches

Analysis of wake around DARPA Suboff including self-propulsi...

引用

2022 Ivannikov Open Conference, ISPRAS 2022

作者： Wang, Xinyu Zhang, Qin Qin, Yingfan Wang, Tianyuan Wang, Zhicheng Epikhin, Andrey Ocean University of China College of Engineering Qingdao China Programming of the Ras Ivannikov Institute for System Moscow Russia

ISBN: (纸本)9798350398533

The Delayed Detached Eddy Simulation(DDES) of the DARPA Suboff, including self-propulsion with sail, fins and seven-blade propeller, is performed at high Reynolds number and pitch-angle inflow besed OpenFOAM, the mixing evolution and energy performance of the vortex structure have been focused. Computations of Suboff are validated against reference data, simulation and experiment of the propeller are carried out simultaneously under the condition of high blade load (J = 0). The time-averaged, phase-averaged, instantaneous vorticity fields and power spectral density (PSD) around the submarine including self-propulsion, are reported under oblique inflow with drift angles of 0°, 7° and 15°. Computations demonstrate that the oblique inflow will deflect and attenuate the horseshoe vortex behind the sail resulting in no longer interacting with fin. Besides, it will speed up the process of the vortex dissipation to a certain extent. With the increase of inflow angle, the hydrodynamic performance of the propeller will have an increasing trend, but the trend will decrease. The fin-sheet vortex accelerates the instability process of the leeward wake, moreover, the horizontal fin-sheet vortex and propeller wake will deflect leeward with the inflow-angle increasing, varying the area of interaction. It is worth noting that the tip-vortex induced by leeward fin is attracted to wake, resulting in the abnormal wake under 15° inflow, and drift opposite to the rotation direction of the propeller. © 2022 IEEE.

关键词： Wakes

来源：评论

学校读者我要写书评

暂无评论

Test Derivation against Timed Finite State Machines with Output Timed Guards

Test Derivation against Timed Finite State Machines with Out...

引用

2023 International Russian Automation Conference, RusAutoCon 2023

作者： Tvardovskii, Aleksandr Yevtushenko, Nina National Research Tomsk State University Tomsk Russia Ivannikov Institute for System Programming of the Ras Moscow Russia

ISBN: (纸本)9798350345551

Approaches based on state machines are widely used to obtain guaranteed error coverage tests for discrete event systems. Many modern systems have time-dependent components, and thus, state machines are extended with clock variables and various time aspects, i.e. test synthesis methods must be adapted to new timed models. In this article, we consider Finite State Machines (FSM) with timeouts, input and output timed guards and show, how a test suite with guaranteed error coverage can be derived based on this model. The fault model and test derivation methods are based on the abstraction of the timed machine, i.e., on the FSM that describes the relevant aspects of the timed behavior. We consider two strategies for obtaining complete test suites under the assumption that the specification can be non-deterministic. © 2023 IEEE.

关键词： Discrete event simulation

来源：评论

学校读者我要写书评

暂无评论

Large-scale YouTube-8M video understanding with deep neural networks

arXiv

引用

arXiv 2017年

作者： Akopyan, Manuk Khashba, Eshsou Institute for System Programming

Video classification problem has been studied many years. The success of Convolutional Neural Networks (CNN) in image recognition tasks gives a powerful incentive for researchers to create more advanced video classification approaches. As video has a temporal content Long Short Term Memory (LSTM) networks become handy tool allowing to model long-term temporal clues. Both approaches need a large dataset of input data. In this paper three models provided to address video classification using recently announced YouTube-8M large-scale dataset. The first model is based on frame pooling approach. Two other models based on LSTM networks. Mixture of Experts intermediate layer is used in third model allowing to increase model capacity without dramatically increasing computations. The set of experiments for handling imbalanced training data has been conducted. Copyright © 2017, The Authors. All rights reserved.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

How To Classify Document Segments Using Graph Based Representation and Neural Networks

How To Classify Document Segments Using Graph Based Represen...

引用

2024 Ivannikov Memorial Workshop, IVMEM 2024

作者： Kopylov, Daniil Mikhaylov, Andrey Matrosov Institute for System Dynamics and Control Theory of Siberian Branch of the Russian Academy of Sciences Irkutsk Russia Ivannikov Institute for System Programming of the Russian Academy of Sciences Moscow Russia

ISBN: (纸本)9798331504663

Bottom-up segmentation approaches, where words are combined into lines, lines into paragraphs, etc., are well-known nowadays. In this work, we use this principle for classification of segments. The features of the blocks that make up a segment and their relationships can be sufficient for classification. Our approach combines a specific choice of features that increases explainability by employing the generalizing ability of neural networks. This work presents an intermediate result, but, nevertheless, the 0.793 classification accuracy on the part of PubLayNet indicates the competitive potential of the approach. It is expected that, similar to the transition from MLP to CNN for MNIST, the transition from vector representation at our stage to GCN at the planned stage will allow us to achieve good results in document classification. © 2024 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Implementation of POD and DMD methods in Apache Spark framework for simulation of unsteady turbulent flow in the model combustor 7

Implementation of POD and DMD methods in Apache Spark framew...

引用

7th European Congress on Computational Methods in Applied Sciences and Engineering, ECCOMAS Congress 2016

作者： Kalugin, Michael D. Strijhak, Sergei V. Institute for System Programming Russian Academy of Sciences Alexander Solzhenitsyn st. 25 Moscow109004 Russia

ISBN: (纸本)9786188284401

The paper is devoted to modelling and analysis of unsteady turbulent flow in a model combustor (channel) using LES (Large Eddy Simulation). Simulations were provided for 2D and 3D cases on different grids of a flow in a channel with rearward facing step. The calculation of a flow in a channel was performed on high performance cluster, using the new approach in Apache Spark framework and POD, DMD data processing algorithms. First 4 dynamic modes were defined.

关键词： Dynamic mode decomposition

来源：评论

学校读者我要写书评

暂无评论

Symbolic Security Predicates: Hunt Program Weaknesses

Symbolic Security Predicates: Hunt Program Weaknesses

引用

2021 Ivannikov Ispras Open Conference, ISPRAS 2021

作者： Vishnyakov, Alexey Logunova, Vlada Kobrin, Eli Kuts, Daniil Parygina, Darya Fedotov, Andrey Ivannikov Institute For System Programming Of The Ras Russia Lomonosov Moscow State University Russia Moscow Institute Of Physics And Technology Moscow Russia

ISBN: (纸本)9781665423304

Dynamic symbolic execution (DSE) is a powerful method for path exploration during hybrid fuzzing and automatic bug detection. We propose security predicates to effectively detect undefined behavior and memory access violation errors. Initially, we symbolically execute program on paths that don't trigger any errors (hybrid fuzzing may explore these paths). Then we construct a symbolic security predicate to verify some error condition. Thus, we may change the program data flow to entail null pointer dereference, division by zero, out-of-bounds access, or integer overflow weaknesses. Unlike static analysis, dynamic symbolic execution does not only report errors but also generates new input data to reproduce them. Furthermore, we introduce function semantics modeling for common C/C++ standard library functions. We aim to model the control flow inside a function with a single symbolic formula. This assists bug detection, speeds up path exploration, and overcomes overconstraints in path predicate. We implement the proposed techniques in our dynamic symbolic execution tool Sydr. Thus, we utilize powerful methods from Sydr such as path predicate slicing that eliminates irrelevant constraints. We present Juliet Dynamic to measure dynamic bug detection tools accuracy. The testing system also verifies that generated inputs trigger sanitizers. We evaluate Sydr accuracy for 11 CWEs from Juliet test suite. Sydr shows 95.59% overall accuracy. We make Sydr evaluation artifacts publicly available to facilitate results reproducibility. © 2021 IEEE.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：