Supercomputing

Name: Supercomputing | 5th Russian Supercomputing Days, RuSCDays 2019, Moscow, Russia, September 23-24, 2019, Revised Selected Papers
Brand: Springer
Price: 96.29 EUR
Availability: OnlineOnly

5th Russian Supercomputing Days, RuSCDays 2019, Moscow, Russia, September 23-24, 2019, Revised Selected Papers

Vladimir Voevodin Sergey Sobolev(Herausgeber*in)

Springer (Verlag)

Erschienen am 9. Dezember 2019

XVIII, 747 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-030-36592-9 (ISBN)

96,29 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Contents
Parallel Algorithms
A Highly Parallel Approach for Solving Computationally Expensive Multicriteria Optimization Problems
1 Introduction
2 Statement of a Multicriteria Optimization Problem
3 Parallel Computations for Solving the Multicriteria Global Optimization Problems
3.1 General Scheme of Parallel Computations
3.2 Accumulation and Reuse of Calculated Optimization Data
3.3 Parallel Computations on Multiprocessors with Shared Memory
4 Results of Numerical Experiments
5 Conclusion
References
Data-Parallel High-Precision Multiplication on Graphics Processing Units
1 Introduction
2 Background
2.1 Residue Number System
2.2 High-Precision Data Representation
3 Parallel High-Precision Multiplication
3.1 Description of the Algorithm
3.2 Data-Parallel GPU Implementations
4 Performance Results
5 Conclusion
References
Efficiency of Basic Linear Algebra Operations on Parallel Computers
1 Introduction
2 Computational Models and Efficiency Estimates
2.1 Shared Memory Model
2.2 Parallel Computation Model for Distributed Memory Computers
3 Examples of Arithmetic Operations
4 Numerical Experiments
5 Conclusion
References
Explicit-Implicit Schemes for Parallel Solving of the Suspension Transport Problems in Coastal Systems
1 Introduction
2 Continuous 3D Model of Diffusion-Convection-Aggregation of Suspensions
3 Constructing an Explicit-Implicit Scheme for the Suspension Transport Problem
4 The Optimal Value of the Weight Parameter in the Scheme with Weights for the Diffusion-Convection Equation
5 Description of the Parallel Algorithm
6 The Results of a Numerical Experiment
7 Conclusion
References
GPU Implementation of a Stencil Code with More Than 90% of the Peak Theoretical Performance
1 Introduction
2 Algorithm
2.1 Recursive Domain Decomposition with Halo
2.2 Performance Model
3 Implementation
3.1 Computation Kernel
3.2 CUDA-Block Synchronization
3.3 FMA Intrinsics
4 Conclusion
References
Memory-Optimized Tile Based Data Structure for Adaptive Mesh Refinement
1 Introduction
2 Data Structure
3 Mesh Traversal and Access to Neighbors
4 Mesh Adaptation
5 Data Serialization
6 Performance Benchmark
7 Conclusion
References
Multithreaded Multifrontal Sparse Cholesky Factorization Using Threading Building Blocks
Abstract
1 Introduction
2 Multifrontal Method Overview
3 Task-Based Two-Level Dynamic Parallel Algorithm
4 Exploiting Parallelism Using the Threading Building Blocks Library
5 Numerical Results
5.1 Computational Infrastructure and Test Problems
5.2 Results and Discussion
6 Conclusion
References
Optimal Packings of Congruent Circles on a Square Flat Torus as Mixed-Integer Nonlinear Optimization Problem
1 Introduction
2 Formulation as Global Optimization MINLP
3 Case Study: Solver SCIP, N8
4 Case Study: Solver ParaSCIP, N = 8, 9
5 Conclusion
References
Parallel Global Optimization for Non-convex Mixed-Integer Problems
1 Introduction
2 Global Optimization Algorithm and Dimension Reduction
3 Parallel Algorithm for Mixed-Integer Problems
4 Results of Experiments
5 Conclusion
References
Parallel Ray Tracing Algorithm for Numerical Analysis of Laser Radiation Absorption in a Plasma
1 Introduction
1.1 Problem Statement
2 Ray Tracing Algorithm
2.1 State of the Art in Ray Tracing and Related Works
2.2 Use of Rational Arithmetic
2.3 Parallel Technique
2.4 Scalability
3 Applications
3.1 Post-processing the Simulation Results, Allowing a Direct Comparison with Experimental Data
3.2 Simulation of Laser Light Propagation
4 Conclusion
References
Porting CUDA-Based Molecular Dynamics Algorithms to AMD ROCm Platform Using HIP Framework: Performance Analysis
1 Introduction
2 Literature Review
3 Methods
4 Results
5 Conclusion
References
Solving of Eigenvalue and Singular Value Problems via Modified Householder Transformations on Shared Memory Parallel Computing Systems
Abstract
1 Introduction
2 Used Methods
3 Proposed Solutions
4 Results of Computational Experiments
5 Summary
Acknowledgements
References
Supercomputer Simulation
Analysis of the Influence of Changes in the Exterior Elements of the Car on the Drag Coefficient and the Problem of Its Optimization
1 Introduction
2 Formulation of the Problem
3 Calculation of the Aerodynamic Drag Coefficient
4 Optimization of the Car's Exterior Shape
5 Calculation Results
6 Conclusions
References
Application of High-Performance Computing for Modeling the Hydrobiological Processes in Shallow Water
1 Introduction
2 Problem Statement
3 Solution Method of Model Problems
4 Parallel Implementation
4.1 Algorithm 1
4.2 Algorithm 2
4.3 Parallel Implementation on Graphic Accelerator
5 Program Complex Description
6 Results of Numerical Experiments
7 Conclusion
References
Computer Simulation of Endothermic Decomposition of High-Energy Substances CxHyNmOn. Substance Composition Optimization
Abstract
1 Simulation of High-Energy Systems and Their Decomposition Processes
1.1 Fuel in Ramjet
1.2 Thermal Decomposition of Fuel
2 Quantum Chemical Simulation
3 Details of Computational Calculations
4 Conclusion
References
Coupling of PDE and ODE Solvers in INMOST Parallel Platform: Application to Electrophysiology
1 Introduction
2 Electrophysiology Model
3 Numerical Methods
4 Numerical Simulations
5 Discussions and Conclusions
References
Digital Rock Modeling of a Terrigenous Oil and Gas Reservoirs for Predicting Rock Permeability with Its Fitting Using Machine Learning
Abstract
1 Introduction
1.1 Method and Theory
1.2 Model Developing
2 Method
2.1 Related Works
2.2 Convolutional Neural Networks (CNN)
2.3 Convolutional Layer
2.4 Pooling Layers
2.5 Fully Connected Layers
2.6 Convolutional Autoencoder
2.7 TensorFlow
2.8 Proposed Methods
2.9 Machine Learning Experiment Results
3 Conclusions
Acknowledgements
References
Digital Twin of the Seismogeological Object: Building and Application
Abstract
1 Introduction
2 Model Building
2.1 Skeleton of the Model
2.2 Simulation of 3D Geological Faults
3 3D Seismic Simulation
3.1 Presentation of the Model
3.2 Hardware and Software Used for Simulation
3.3 Justification of Equipment Selection
4 Scattering Imaging of Microheterogeneities
5 Conclusions
Acknowledgments
References
GPU-Based Discrete Element Modeling of Geological Faults
1 Introduction
2 Discrete Elements Method
2.1 Forces Computation
2.2 Time Integration
2.3 Boundary Conditions
2.4 Output Parameters
3 Implementation of the Algorithm
4 Numerical Experiments
4.1 Clustering of the Results
5 Conclusions
References
High Performance Parallel Simulations of Subsurface Radar Sounding of Celestial Bodies
1 Introduction
2 Simulations of Radar Pulses in Plane-Parallel Layered Media
3 Phase Screen Technique for Simulation of Wave Propagation
4 Discussion of the Results
5 Conclusions and Remarks
References
High-Performance Hybrid Computing for Bioinformatic Analysis of Protein Superfamilies
1 Introduction
2 Methods
2.1 The Hardware
2.2 Overview of the Mustguseal Protocol
2.3 Step 1: Structure Similarity Search
2.4 Step 2: 3D-Alignment
2.5 Step 3: Sequence Similarity Search
2.6 Step 4: Structure-Guided Sequence Alignment
2.7 Multiple Alignment of the PLP-Dependent Enzymes
3 Results
4 Discussion and Conclusions
References
Improving Parallel Efficiency of a Complex Hydrogeological Problem Simulation in GeRa
1 Introduction
2 The Model of Projected RW Deep Geological Disposal
3 Optimization of Linear Solver Parameters
4 Parallel Efficiency Analysis
5 Conclusion
References
Morphing and Wave Perturbations in Distinguishing Molecular Clouds' Collision
Abstract
1 Introduction
2 Problem Setup
2.1 Eulerian Description and Numerical Realization
2.2 Modeling Assumptions and Parameters Used
3 Morphological Peculiarities of MCs Head-on Collision
3.1 Clouds Matter Interplay During Collision
3.2 Instabilities in a Bow-Shock Layer and Post-collision Shell Shaping
4 Conclusions
Acknowledgements
References
Numerical Experiments with Digital Twins of Core Samples for Estimating Effective Elastic Parameters
1 Introduction
2 Statement of the Problem
3 The Method
4 The Algorithm for Determining the Components of the Tensor S
5 Numerical Solution to the Static Elasticity Problem
6 Parallel Implementation
6.1 MPI/OpenMP Parallelization
6.2 Parallelization Approach with Coarray Fortran
7 Numerical Experiments
7.1 Validation of the Numerical Algorithm
7.2 Comparison of MPI, MPI+OpenMP and Coarray Fortran
8 Conclusion
References
Orange Carotenoid Protein Absorption Spectra Simulation Using the Differential Evolution Algorithm
Abstract
1 Introduction
2 Materials and Methods
2.1 Sample Preparation and Characterization
2.2 Differential Evolution
2.3 Programming Details
3 Theory of the Multimode Brownian Oscillator Model
4 Results
5 Conclusions
Acknowledgement
References
Parallel Dynamic Mesh Adaptation Within INMOST Platform
1 Introduction
2 Internal Representation
3 Parallel Mesh Modification
3.1 Resolve Modified Elements
3.2 Equilibrate Shared Elements
3.3 Reconstruct Ghost Layers
4 Data Transfer
5 Mesh Balancing
6 General Mesh Refinement and Coarsening
7 Numerical Tests
8 Conclusion
References
Parallel Implementation of Coupled Wave and Bottom Deposit Transportation Models to Simulate Surface Pollution Areas
1 Introduction
2 Statement of Wave Hydrodynamics Problem
3 Discrete Model of Shallow Water Reservoirs Hydrodynamics
4 Continuous 3D Model of Diffusion-Convection-Aggregation of Suspensions
5 2D Model of Sediment Transport
6 Parallel Version of Algorithm for Solving Grid Equations
7 Results of Numerical Experiments
8 Conclusion
References
Recovery of the Permittivity of an Anisotropic Inhomogeneous Body in a Rectangular Waveguide
1 Introduction
2 Direct Diffraction Problem
3 Inverse Diffraction Problem
4 Parallel Algorithm and Issue of Saving Matrix Coefficients
5 Conclusion
References
Relativistic Hydrodynamics Modeling by Means Adaptive Nested Mesh on IBM Power 9
1 Introducation
2 The Numerical Method
2.1 The Equations
2.2 The Primitive Variables Recovery
2.3 The Relativistic Hydrodynamics Solver
3 The Code Architecture
4 The Research of Code on IBM Power 9 Nodes
5 The Numerical Expriments of Relativistic Hydrodynamics of Jet Galaxy Formation
6 Conclusion
References
Search for Approaches to Supercomputer Quantum-Chemical Docking
Abstract
1 Introduction
2 Materials and Methods
2.1 Quasi-Docking
2.2 FLM Program
2.3 PM7 Method
2.4 COSMO Solvent Model
2.5 Protein-Ligand Binding Enthalpy
2.6 Test Set of Protein-Ligand Complexes
2.7 Energy Indexes: INN
3 Results
3.1 Positioning Accuracy
3.2 Binding Enthalpy
3.3 Discussion of the Results
4 Conclusions
Acknowledgements
References
Simulation of ``ExoMars'' Spacecraft Landing on the Surface of Mars Using Supercomputer Technologies
1 Introduction
2 Problem Statement
2.1 Mathematical Model for Gas Flow in Nozzles and Jets
2.2 Lift Loss Effect
2.3 Erosion
2.4 Particle Motion
3 Numerical Implementation and Acceleration of Calculations
4 Numerical Results and Discussion
4.1 The Results of Jet Interaction with Surfaces
4.2 The Results of Lift Loss Effect
4.3 The Results of Land Surface Erosion
4.4 The Results of Particle Motion
5 Conclusions
References
Solving Inverse Problems of Ultrasound Tomography in a Nondestructive Testing on a Supercomputer
Abstract
1 Introduction
2 Formulation of the Inverse Problem of Ultrasonic Tomography in Nondestructive Testing
3 Numerical Tests of the Inverse Problem of Ultrasonic Tomography in Nondestructive Testing
4 Reconstruction of the Velocity Structure from Experimental Data
5 Conclusion
Acknowledgement
References
Supercomputer Modelling of Spatially-heterogeneous Coagulation using MPI and CUDA
1 Introduction
2 Advection-Coagulation Equation and Numerical Method
2.1 Advection-Coagulation Equation
2.2 Discretization and Algorithm
2.3 Numerical Scheme for Advection Process
2.4 Numerical Approximation of Coagulation
2.5 Full Algorithm
3 Parallel Implementation
3.1 CPU Parallelization
3.2 Adding GPU Parallelism
4 Results
References
Supercomputing the Seasonal Weather Prediction
1 Introduction
2 Why Supercomputing?
3 Coupled Model Configuration
4 Computational Characteristics of the Coupled Model
5 Conclusion
References
The Creation of Intelligent Support Methods for Solving Mathematical Physics Problems on Supercomputers
Abstract
1 Introduction
2 Basic Principles of Building IS System
3 Designing the Knowledge Base
4 An Example of Using the Intelligent Support System for Solving Astrophysics Problems
5 Conclusions
Acknowledgements
References
The Parallel Implementation of the Adaptive Mesh Technology in Poroelasticity Problems
Abstract
1 Introduction
2 Some Aspects of Using Adaptive Grids
3 The Algorithm of Adaptive Mesh Construction
4 Using an Adaptive Grid for Geomechanical Problems
5 The Statement of Poroelasticity Problem
6 The Parallel Implementation of a Finite Element Method
7 Conclusion
Acknowledgements
References
The Simulation of 3D Wave Fields in Complex Topography Media
1 Introduction
2 Statement of the Problem
2.1 Mathematical Model in Cartesian Coordinates
2.2 Mathematical Model in Generalized Coordinates
3 Constructing a Curvilinear Mesh
4 Finite Difference Approximation
5 Parallel Algorithm and Its Realization
6 Numerical Simulation and Scalability Tests
7 Conclusion
References
Three-Dimensional Ultrasound Tomography: Mathematical Methods and Experimental Results
1 Introduction
2 Formulation of the Inverse Problem of 3D Ultrasound Tomography and Its Solution Methods
3 GPU Algorithm for Solving the Inverse Problem of Wave Tomography Using Experimental Data
3.1 Flowchart of the Algorithm
3.2 GPU Implementation of the Iterative Gradient Method of Minimizing the Residual Functional
3.3 Multistage Algorithm for Solving Inverse Problem of Wave Tomography
4 Results of 3D Sound Speed Image Reconstruction Using Experimental Data
5 Conclusion and Discussion
References
Validation of the Regional Climate Model for the South of Russia
1 Introduction
2 Components and Structure of RegCM 4.5
3 Testing the RegCM 4.5 Radiative Code
4 Results and Experience of the Supercomputer Modeling
5 Conclusion
References
HPC, BigData, AI: Architectures, Technologies, Tools
An Experimental Study of Deep Neural Networks on HPC Clusters
Abstract
1 Introduction
1.1 Related Works
1.2 Scope
2 Description of Computing Platforms
2.1 Cluster Polus Architecture
2.2 Hardware Configuration of the Lomonosov-2 Supercomputer Section
2.3 Software
3 DNN Models and Experimental Methodology
4 Experimental Results on Polus
4.1 Serial Implementation of Mobilenet V2
4.2 Parallel Implementation of DNN-Models on Polus
5 Experiment Results on Lomonosov-2
5.1 Serial Implementation of Mobilenet V2
5.2 Parallel Implementation
5.3 Comparison of SqueezeNet for Batch Size 24: Lomonosov-2 - Polus
6 Conclusion
References
Artificial Intelligence Problems in Mathematical Modeling
1 Introduction
2 Features of Intellectualization of Technological Stages of Modeling
2.1 Tasks of Geometric and Functional Modeling
2.2 The Intellectialization of the Grid Generation
2.3 Automation of Approximation Procedures
2.4 Automatic Construction of the Algebraic Algorithms
3 Methods and Tools of Artificial Intelligence for Mathematical Modeling
4 Conclusion
References
Aspect-Oriented Set@l Language for Architecture-Independent Programming of High-Performance Computer Systems
Abstract
1 Introduction
2 Aspect-Oriented Approach to Programming of Computer Systems
3 Description of Algorithm Parallelization in Set@l
4 Indefinite Collections in Set@l
5 Program of Lower-Upper Decomposition in Set@l
6 Conclusions
References
Bridging the Gap Between Applications and Supercomputing: A New Master's Program in Computational Science
Abstract
1 Introduction
2 Related Work
3 Main Principles
4 Curriculum Structure
5 Courses Description
5.1 Term I
5.2 Term II
5.3 Term III
6 Implementation
7 Conclusion
Acknowledgements
References
DLI: Deep Learning Inference Benchmark
1 Introduction
2 Related Work
3 Benchmark Architecture
4 Implementation
5 Experiments
5.1 Hardware and Software
5.2 Models
5.3 Datasets
5.4 Experiment Types
5.5 Measurements
5.6 Performance Analysis
6 Conclusion
References
Educational Course ``Introduction to Deep Learning Using the Intel neon Framework''
1 Introduction
2 Related Work
3 Main Principles
4 Course Structure
4.1 Course Description
4.2 Course Theory
4.3 Course Practice
5 Implementation
6 Conclusion
References
Evaluation of Intel Memory Drive Technology Performance for Computational Astrophysics
Abstract
1 Introduction
2 Intel Memory Drive Technology (IMDT)
3 Mathematical Model and Numerical Method
3.1 Mathematical Model
3.2 Numerical Method
4 Hardware and Software Configuration
5 Benchmarking
6 Conclusion
Acknowledgments
References
Extended Routing Table Generation Algorithm for the Angara Interconnect
1 Introduction
2 Preliminaries
3 Angara Interconnect Routing
3.1 Direction-Order Routing and Dirbit Routing
3.2 First Step/Last Step
4 Extended Direction Order Routing
4.1 Routing Table Generation Algorithm
5 Algorithm Evaluation
6 Conclusion
References
High-Performance Solution of the Two-Class SVM Problem for Big Data Sets by the Mean Decision Rule Method
Abstract
1 Introduction
2 Mean Decision Rule (MDR) Method
2.1 The Main Idea of the MDR Method
2.2 Sequential MDR Algorithm
2.3 Parallel MDR Algorithm
3 Loading Data Optimization
3.1 The Input Data Format
3.2 Data Strategies
4 Experimental Setup
4.1 Data Description
4.2 Study of the Parallel MDR with the Traditional Data Strategy
4.3 Study of the Parallel MDR with the Optimized Data Strategy
4.4 Influence of Parameters of the Parallel MDR on Its Performance and Accuracy
5 Conclusions
Acknowledgements
References
Performance and Scalability of Materials Science and Machine Learning Codes on the State-of-Art Hybrid Supercomputer Architecture
1 Introduction
2 New Supercomputer in NRU HSE
3 Interconnect Benchmarks
4 Molecular Dynamics Benchmarks with Gromacs
5 Electronic Structure Calculations Benchmarks with CP2K
6 Electronic Structure Calculations Benchmarks with VASP
7 Machine Learning: Performance and Collocation Effects
8 Conclusions
References
Performance of the Particle-in-Cell Method with the Intel (Broadwell, KNL) and IBM Power9 Architectures
1 Introduction
2 The Particle-in-Cell Method. The Main Features of Algorithms
3 Code Optimization
3.1 General Remarks
3.2 Optimization Options
3.3 Tests of 3D Code
4 Performance Tests of the 2D Code with the Intel and IBM Architectures
5 Scalability of Computations
6 Conclusion
References
Software Development Tools for FPGA-Based Reconfigurable Systems Programming
Abstract
1 Introduction
2 Software Development Tools for Next-Generation Reconfigurable Computer Systems
3 The Methodology of Transformation of a Sequential C-Program into a Parallel COLAMO-Program
3.1 Transformation of a C-Program into an Absolutely Parallel COLAMO-Program
3.2 Analysis of Information Dependences of the Source Program, Splitting of Scalar Variables and Stretching of Arrays by Iterations
3.3 Conversion into the Parallel-Pipeline Form, Separation of Arrays and Loops into Vector and Stream Components
3.4 Feedback Implementation
3.5 Transformation into the Scalable Parallel-Pipeline Form
3.6 Generation of a Computing Structure
4 Translation Results of Several Application Tasks
5 Conclusion
References
Software Q-system for the Research of the Resource of Numerical Algorithms Parallelism
1 Introduction
2 Theoretical Foundation of Q-system
2.1 Method of Constructing of Q-determinant of Algorithm
2.2 Method of Obtaining of Q-effective Implementation of Algorithm
2.3 Method of Calculating Characteristics of a Parallel Complexity of Q-effective Implementation of Algorithm
2.4 Method of Comparing the Characteristics of the Parallel Complexity of Q-effective Implementations of Algorithms
3 Program Realization of Software Q-system
3.1 Subsystem for the Creation of Q-determinants of Numerical Algorithms
3.2 Subsystem for the Calculating of the Resource of Numerical Algorithms Parallelism
4 Results of Trial Operation of the Q-system
5 Conclusion
References
The Role of Student Projects in Teaching Machine Learning and High Performance Computing
1 Introduction
2 Related Work
3 The Structure of the Master Program
4 Project Workshop
5 Examples of the Student Projects
5.1 The Tenants Requests Classification System
5.2 Obscenity Detection
5.3 Prediction of the Government Procurement Results
6 On the Convergence of Machine Learning and High Performance Computing in the Projects
7 Discussion
8 Conclusion
References
Distributed and Cloud Computing
Budget and Cost-Aware Resources Selection Strategy in Cloud Computing Environments
Abstract
1 Introduction
2 Problem Definition
3 Related Works
4 Resource Selection Algorithm
4.1 CloudAuction
4.2 Resource Selection Procedure
5 Experimental Analysis
5.1 Experimental Setup
5.2 Simulation System
5.3 Experimental Results
6 Conclusion and Future Work
Acknowledgments
References
Building an Algorithmic Skeleton for Block Data Processing on Enterprise Desktop Grids
Abstract
1 Introduction
2 Related Work
3 The Actor Model of Algorithmic Skeletons for Desktop Grid Applications
4 Specification of Bag-of-Tasks Skeleton
5 Specification of Asynchronous Round-Robin Tournament Skeleton
6 Experimental Study of Asynchronous Round-Robin Tournament Skeleton
7 Discussion
8 Conclusions
References
Roadmap for Improving Volunteer Distributed Computing Project Performance
Abstract
1 Introduction
2 Additional Tasks for Desktop Grid
3 Roadmap for VDC Project
4 Methods of Obtaining Data for Multivariate Assessment of VDC Projects
4.1 Questionnaire for Multiparameter Assessment of VDC Projects
4.2 Weights of Characteristics
4.3 Multiparameter Evaluation of Gerasim@home VDC Project in 2016
4.4 Multiparameter Evaluation of Gerasim@home VDC Project in 2018
5 Conclusions
Acknowledgements
References
SAT-Based Cryptanalysis: From Parallel Computing to Volunteer Computing
1 Introduction
2 Cryptanalysis Problems Solved in SAT@home
3 New SAT-Based Cryptanalysis Problems
4 Preparation of Experiment for SAT@home
4.1 Analysis via Multithreaded Solvers
4.2 Finding Decompositions Sets for Hard Instances
4.3 Details of Future SAT@home Experiments
5 Conclusions
References
Supporting Efficient Execution of Workflows on Everest Platform
1 Introduction
2 Everest Platform
3 Workflow Execution via Everest Client Library
4 Workflow Runner Service
4.1 Workflow Description
4.2 Workflow Execution
4.3 Data Transfer Optimizations
4.4 Scheduling Optimization
5 Experimental Evaluation
6 Conclusion and Future Work
References
Start-up and the Results of the Volunteer Computing Project RakeSearch
1 Introduction
1.1 Orthogonal Diagonal Latin Squares
1.2 Desktop Grids
1.3 Search for Orthogonal Diagonal Latin Squares Using Desktop Grids
2 Setup of the BOINC-Based Desktop Grid
2.1 Server Infrastructure
2.2 An Approach to Fast Discovery of Orthogonal Diagonal Mates
2.3 Algorithm of Search for Row-Permutational ODLSs
3 RakeSearch Project
4 Results of the Project
5 Conclusion
References
Use of a Desktop Grid to Effectively Discover Hits in Virtual Drug Screening
1 Introduction
2 Related Work
3 Mathematical Model
4 Computational Process
5 Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken