| |
|
|
|
|
|
|
|
|
1. |
Record Nr. |
UNINA9910254998903321 |
|
|
Titolo |
Supercomputing : Second Russian Supercomputing Days, RuSCDays 2016, Moscow, Russia, September 26–27, 2016, Revised Selected Papers / / edited by Vladimir Voevodin, Sergey Sobolev |
|
|
|
|
|
|
|
Pubbl/distr/stampa |
|
|
Cham : , : Springer International Publishing : , : Imprint : Springer, , 2016 |
|
|
|
|
|
|
|
|
|
ISBN |
|
|
|
|
|
|
Edizione |
[1st ed. 2016.] |
|
|
|
|
|
Descrizione fisica |
|
1 online resource (XV, 370 p. 166 illus.) |
|
|
|
|
|
|
Collana |
|
Communications in Computer and Information Science, , 1865-0937 ; ; 687 |
|
|
|
|
|
|
|
|
Disciplina |
|
|
|
|
|
|
Soggetti |
|
Computer simulation |
Computer programming |
Computer science - Mathematics |
Computers |
Numerical analysis |
Algorithms |
Computer Modelling |
Programming Techniques |
Mathematical Applications in Computer Science |
Computer Hardware |
Numerical Analysis |
|
|
|
|
|
|
|
|
Lingua di pubblicazione |
|
|
|
|
|
|
Formato |
Materiale a stampa |
|
|
|
|
|
Livello bibliografico |
Monografia |
|
|
|
|
|
Nota di contenuto |
|
Intro -- Preface -- Organization -- Contents -- The Present of Supercomputing: Large Tasks Solving Experience -- Accelerating Assembly Operation in Element-by-Element FEM on Multicore Platforms -- 1 Introduction -- 2 Element-by-Element FEM on Shared Memory Platforms -- 3 FEM Algorithms with Layer-by-Layer Partitioning -- 4 Experimental Platforms and Their Memory Features -- 5 Scalability Model for Multi-processor Nodes -- 6 Experimental Results -- 7 Conclusion -- References -- Block Lanczos--Montgomery Method with Reduced Data Exchanges -- 1 Introduction -- 2 Description of the Improved Lanczos--Montgomery Algorithm -- 2.1 The Lanczos |
|
|
|
|
|
|
|
|
|
Algorithm for Linear Systems over Finite Fields -- 2.2 Parallel Computing in the Improved Lanczos--Montgomery Method -- 3 Parallel Complexity Analysis for the Improved Lanczos--Montgomery Method -- 3.1 Complexity Estimate for the Lanczos Method over Large Fields -- 3.2 Complexity Estimate for the Montgomery Method over F2 -- 3.3 Parallel Complexity Analysis for the Lanczos--Montgomery-like Methods -- 4 Numerical Experiments -- References -- ChronosServer: Fast In Situ Processing of Large Multidimensional Arrays with Command Line Tools -- Abstract -- 1 Introduction -- 2 In-Situ Raster Data Processing -- 2.1 State-of-the-Art -- 2.2 In-Situ Approach Benefits -- 3 ChronosServer Architecture -- 3.1 Raster Data Model: Abstracting from Files, Their Locations and Formats -- 3.2 Cluster Orchestration -- 4 New Delegation Approach -- 4.1 ChronosServer Raster Data Processing Commands and Their Distributed Execution -- 4.2 Distributed Apply-Combine-Finally Execution Scheme (Under Development) -- 4.3 Benefits of the Proposed Delegation Approach -- 5 Performance Evaluation -- 5.1 Test Raster Data and Experimental Setup -- 5.2 SciDB Data Import and ChronosServer Data Discovery -- 5.3 Simple Statistics. |
5.4 User-Defined Arithmetic Expressions -- 5.5 Multidimensional Chunking -- 6 Conclusion -- Acknowledgements -- A Appendix. ChronosServer Queries -- B Appendix. SciDB Queries -- References -- Dynamics of Formation and Fine Structure of Flow Pattern Around Obstacles in Laboratory and Computational Experiment -- Abstract -- 1 Introduction -- 2 Governing Equations, Basic Scales and Simulation Conditions -- 3 Laboratory Modeling of Flows Around a Plate -- 4 Method for Numerical Simulation of Flow Around Obstacles -- 5 Calculation Results -- 5.1 The Structure of Diffusion-Induced Flow on a Motionless Plate -- 5.2 Diffusion-Induced Flow on a Wedge -- 5.3 Flow Around a Rectangular Strip -- 6 Conclusion -- Acknowledgements -- References -- EnOI-Based Data Assimilation Technology for Satellite Observations and ARGO Float Measurements in a High Resolution Global Ocean Model Using the CMF Platform -- Abstract -- 1 Introduction -- 2 The Compact Computing Platform for Modeling CMF 3.0 and Service DAS -- 2.1 PGAS-Communicator -- 2.2 The New Architecture of Coupling Model -- 2.3 Coupler: Interpolation -- 3 Description of DAS Service -- 3.1 EnOI Method: Basic Equations -- 3.2 Features of the Parallel Implementation of EnOI Method -- 3.3 Service DAS Testing -- 4 Conclusion -- References -- Experience of Direct Numerical Simulation of Turbulence on Supercomputers -- Abstract -- 1 Introduction -- 2 NTS Code -- 2.1 NTS Code Numerics -- 2.2 Algorithm Parallelization -- 3 Parallelization Efficiency -- 4 Description of the Production Simulation and Its Results -- 4.1 Problem Size, Computational Productivity, and Consumed Computational Resources -- 4.2 Major Physical Results of the Simulation -- 5 Concluding Remarks -- Acknowledgments -- References -- GPU-Accelerated Molecular Dynamics: Energy Consumption and Performance -- 1 Introduction -- 2 Related Work. |
3 Software and Algorithms -- 3.1 A Peak Load Benchmark: Empirical Roofline Toolkit -- 3.2 Classical Molecular Dynamics: LAMMPS -- 4 Hardware -- 4.1 Tested Platforms -- 4.2 Energy Consumption Measurement Technique -- 5 Measurements Results -- 5.1 Energy Consumption for the ERT Benchmark -- 5.2 Energy Consumption for the LAMMPS Benchmark -- 6 GPU and DRAM Frequencies Variation Effect -- 7 Summary -- References -- Implementation and Evaluation of the PO-HEFT Problem-Oriented Workflow Scheduling Algorithm for Cloud Environments -- Abstract -- 1 Introduction -- 2 Review of Workflows Scheduling Methods -- 2.1 Independent Tasks and |
|
|
|
|
|
|
|
Workflows Scheduling Algorithms -- 2.2 Algorithms of Resource Utilization Prediction -- 2.3 Workflow Simulation -- 3 Implementation of the PO-HEFT Algorithm -- 4 Simulation of Workflow Execution System -- 4.1 Architecture of the Workflow Execution System -- 4.2 Model of the Executions History Database -- 5 Evaluation of Workflow Scheduling Algorithms -- 5.1 Individual Tasks Prediction Algorithm -- 5.2 PO-HEFT Algorithm -- 6 Conclusion -- Acknowledgements -- References -- Layer-by-Layer Partitioning of Finite Element Meshes for Multicore Architectures -- 1 Introduction -- 2 Layer-by-Layer Partitioning of Finite Element Meshes -- 3 Forming Layers Using the Neighborhood Criterion -- 4 Combining Layers into Subdomains -- 5 Experimental Results -- 6 Conclusion -- References -- Multilevel Parallelization: Grid Methods for Solving Direct and Inverse Problems -- 1 Introduction -- 2 Geometric Decomposition Pattern -- 3 Multicore Threading on Intel Xeon Phi Accelerators -- 4 Vectorization -- 5 Study of Run-Time Energy Efficiency -- 6 Results -- 6.1 Modeling of Wave Propagation Through an Elastic Medium -- 6.2 Modeling of Magneto-Hydrodynamics Turbulence Evolution -- 7 Conclusions -- References. |
Numerical Model of Shallow Water: The Use of NVIDIA CUDA Graphics Processors -- 1 Introduction -- 2 Mathematical and Numerical Models -- 2.1 Basic Equations -- 2.2 Grids System and Matrix of Digital Terrain Elevation -- 2.3 The Numerical Hydrodynamic Scheme -- 3 Parallel Realization of Numerical Model -- 4 Comparison of the Effectiveness for Different GPU -- 5 The Simulation Results -- 6 Conclusion -- References -- Parallel Algorithm for Simulation of Fragmentation and Formation of Filamentous Structures in Molecular Clouds -- Abstract -- 1 Introduction -- 2 Simulation of Shock Wave Collision with Two MCs -- 2.1 Initial Conditions -- 2.2 Parallelization and Performance Optimization -- 2.3 Analysis of MC Forming after Shock Impact -- Acknowledgements -- References -- Parallel Algorithms for a 3D Photochemical Model of Pollutant Transport in the Atmosphere -- Abstract -- 1 Introduction -- 2 Three-Dimensional Predictive Model of Pollutant Transport with Chemical and Photochemical Reactions -- 3 Initial and Boundary Conditions. Deposition and Emission -- 4 Numerical Calculation Method -- 5 Parallelization of the Numerical Algorithm -- 6 Implementation of the Two-Dimensional Decomposition -- 7 Speedup and Efficiency -- 8 Numerical Experiment -- 9 Conclusions -- Acknowledgements -- References -- Parallel Computation of Normalized Legendre Polynomials Using Graphics Processors -- 1 Introduction -- 2 Extended-Range Arithmetic -- 2.1 Basic Algorithms -- 2.2 Implementation of Extended-Range Arithmetic -- 3 Computation of Normalized Legendre Polynomials on CPU and GPU -- 3.1 Computation of Starting Point of Recursion -- 3.2 Developed Software for Computing Legendre Polynomials -- 4 Experimental Results -- 5 Conclusion -- References -- Parallel Software for Simulation of Nonlinear Processes in Technical Microsystems -- 1 Introduction. |
2 Statements of Mathematical Problems -- 2.1 Mathematical Model for the Problem of Gasdynamic Spray -- 2.2 Mathematical Model of Interconnects Degradation -- 3 Numerical Methods, Parallel Algorithms and Programs -- 3.1 Numerical Algorithm for Solving the Problems of Spraying -- 3.2 Numerical Algorithm for Solving the Problems of Interconnects Degradation -- 3.3 Parallel Implementation of Proposed Approach -- 3.4 Program Realization -- 4 Results -- 4.1 Simulation of Gas Flows in Microchannels -- 4.2 Modeling the Processes of Void Formation -- References -- Performance of MD-Algorithms on Hybrid Systems-on-Chip Nvidia Tegra K1 & X1 -- 1 Intoduction -- 2 Literature Review -- 3 Test Setup -- 3.1 Hardware -- 3.2 Software -- 4 Roofline |
|
|
|
|
|
|
|
|
|
Performance Model -- 4.1 Roofline -- 4.2 Theoretic Estimates -- 4.3 Analysis -- 5 Classic Molecular Dynamics and LAMMPS -- 5.1 Molecular Dynamics Method -- 5.2 LAMMPS and Hybrid Architectures -- 5.3 Test Model -- 6 Conclusion -- References -- Revised Pursuit Algorithm for Solving Non-stationary Linear Programming Problems on Modern Computing Clusters with Manycore Accelerators -- 1 Introduction -- 2 Problem Statement -- 3 Description of the Revised Algorithm -- 4 Implementation of Revised Algorithm -- 4.1 Diagram of Main Subroutine -- 4.2 Diagram of Subroutine Calculating Pseudoprojection -- 5 Computing Pseudoprojection on Intel Xeon Phi -- 6 Conclusion -- References -- Solving Multidimensional Global Optimization Problems Using Graphics Accelerators -- 1 Introduction -- 2 Core Global Search Algorithm with Parallel Trials -- 3 Dimension Reduction -- 3.1 Dimension Reduction Using Peano Curves -- 3.2 Nested Optimization Scheme -- 4 Organization of Parallel Computing -- 5 Numerical Experiments -- 6 Conclusion -- References. |
Supercomputer Simulation of Physicochemical Processes in Solid Fuel Ramjet Design Components for Hypersonic Flying Vehicle. |
|
|
|
|
|
|
Sommario/riassunto |
|
This book constitutes the refereed proceedings of the Second Russian Supercomputing Days, RuSCDays 2016, held in Moscow, Russia, in September 2016. The 28 revised full papers presented were carefully reviewed and selected from 94 submissions. The papers are organized in topical sections on the present of supercomputing: large tasks solving experience; the future of supercomputing: new technologies. |
|
|
|
|
|
|
|
| |