High-Performance Computing on Complex Environments

With recent changes in multicore and general-purpose computing on graphics processing units, the way parallel computers are used and programmed has drastically changed. It is important to provide a comprehensive study on how to use such machines written by specialists of the domain. The book provides recent research results in high-performance computing on complex environments, information on how to efficiently exploit heterogeneous and hierarchical architectures and distributed systems, detailed studies on the impact of applying heterogeneous computing practices to real problems, and applications varying from remote sensing to tomography. The content spans topics such as Numerical Analysis for Heterogeneous and Multicore Systems; Optimization of Communication for High Performance Heterogeneous and Hierarchical Platforms; Efficient Exploitation of Heterogeneous Architectures, Hybrid CPU+GPU, and Distributed Systems; Energy Awareness in High-Performance Computing; and Applications of Heterogeneous High-Performance Computing.
* Covers cutting-edge research in HPC on complex environments, following an international collaboration of members of the ComplexHPC
* Explains how to efficiently exploit heterogeneous and hierarchical architectures and distributed systems
* Twenty-three chapters and over 100 illustrations cover domains such as numerical analysis, communication and storage, applications, GPUs and accelerators, and energy efficiency

List of contents

Contributors xxiii
 
Preface xxvii
 
PART I INTRODUCTION 1
 
1. Summary of the Open European Network for High-Performance Computing in Complex Environments 3
Emmanuel Jeannot and Julius Zilinskas
 
1.1 Introduction and Vision 4
 
1.2 Scientific Organization 6
 
1.3 Activities of the Project 6
 
1.4 Main Outcomes of the Action 7
 
1.5 Contents of the Book 8
 
PART II NUMERICAL ANALYSIS FOR HETEROGENEOUS AND MULTICORE SYSTEMS 11
 
2. On the Impact of the Heterogeneous Multicore and Many-Core Platforms on Iterative Solution Methods and Preconditioning Techniques 13
Dimitar Lukarski and Maya Neytcheva
 
2.1 Introduction 14
 
2.2 General Description of Iterative Methods and Preconditioning 16
 
2.3 Preconditioning Techniques 20
 
2.4 Defect-Correction Technique 21
 
2.5 Multigrid Method 22
 
2.6 Parallelization of Iterative Methods 22
 
2.7 Heterogeneous Systems 23
 
2.8 Maintenance and Portability 29
 
2.9 Conclusion 30
 
3. Efficient Numerical Solution of 2D Diffusion Equation on Multicore Computers 33
Matjaz Depolli, Gregor Kosec, and Roman Trobec
 
3.1 Introduction 34
 
3.2 Test Case 35
 
3.3 Parallel Implementation 39
 
3.4 Results 41
 
3.5 Discussion 45
 
3.6 Conclusion 47
 
4. Parallel Algorithms for Parabolic Problems on Graphs in Neuroscience 51
Natalija Tumanova and Raimondas Ciegis
 
4.1 Introduction 51
 
4.2 Formulation of the Discrete Model 53
 
4.3 Parallel Algorithms 59
 
4.4 Computational Results 63
 
4.5 Conclusions 69
 
PART III COMMUNICATION AND STORAGE CONSIDERATIONS IN HIGH-PERFORMANCE COMPUTING 73
 
5. An Overview of Topology Mapping Algorithms and Techniques in High-Performance Computing 75
Torsten Hoefler, Emmanuel Jeannot, and Guillaume Mercier
 
5.1 Introduction 76
 
5.2 General Overview 76
 
5.3 Formalization of the Problem 79
 
5.4 Algorithmic Strategies for Topology Mapping 81
 
5.5 Mapping Enforcement Techniques 82
 
5.6 Survey of Solutions 85
 
5.7 Conclusion and Open Problems 89
 
6. Optimization of Collective Communication for Heterogeneous HPC Platforms 95
Kiril Dichev and Alexey Lastovetsky
 
6.1 Introduction 95
 
6.2 Overview of Optimized Collectives and Topology-Aware Collectives 97
 
6.3 Optimizations of Collectives on Homogeneous Clusters 98
 
6.4 Heterogeneous Networks 99
 
6.5 Topology- and Performance-Aware Collectives 100
 
6.6 Topology as Input 101
 
6.7 Performance as Input 102
 
6.8 Non-MPI Collective Algorithms for Heterogeneous Networks 106
 
6.9 Conclusion 111
 
7. Effective Data Access Patterns on Massively Parallel Processors 115
Gabriele Capannini, Ranieri Baraglia, Fabrizio Silvestri, and Franco Maria Nardini
 
7.1 Introduction 115
 
7.2 Architectural Details 116
 
7.3 K-Model 117
 
7.4 Parallel Prefix Sum 120
 
7.5 Bitonic Sorting Networks 126
 
7.6 Final Remarks 132
 
8. Scalable Storage I/O Software for Blue Gene Architectures 135
Florin Isaila, Javier Garcia, and Jesús Carretero
 
8.1 Introduction 135
 
8.2 Blue Gene System Overview 136
 
8.3 Design and Implementation 138
 
8.4 Conclusions and Future Work 142
 
PART IV EFFICIENT EXPLOITATION OF HETEROGENEOUS ARCHITECTURES 145
 
9. Fair Resource Sharing for Dynamic Scheduling of Workflows on Heterogeneous Systems 147
Hamid Arabnejad, Jorge G. Barbosa, and Frédéric Suter
 
9.1 Introduction

About the author

EMMANUEL JEANNOT is a Senior Research Scientist at INRIA. He received his PhD in computer science from École Normale Supérieur de Lyon. His main research interests are processes placement, scheduling for heterogeneous environments and grids, data redistribution, algorithms, and models for parallel machines. JULIUS ILINSKAS is a Principal Researcher and a Head of Department at Vilnius University, Lithuania. His research interests include parallel computing, optimization, data analysis, and visualization.

Summary

With recent changes in multicore and general-purpose computing on graphics processing units, the way parallel computers are used and programmed has drastically changed. It is important to provide a comprehensive study on how to use such machines written by specialists of the domain.

Product details

• Authors: Jeannot, Emmanuel Jeannot, Emmanuel Zilinskas Jeannot, Julius Zilinskas
• Assisted by: Emmanue Jeannot (Editor), Emmanuel Jeannot (Editor), Zilinskas (Editor), Zilinskas (Editor), Julius Zilinskas (Editor)
• Publisher: Wiley, John and Sons Ltd
• Languages: English
• Product format: Hardback
• Released: 01.07.2014
• EAN: 9781118712054
• ISBN: 978-1-118-71205-4
• No. of pages: 512
• Series: Wiley Series on Parallel and Distributed Computing; Wiley Series on Parallel and D; Wiley Series on Parallel and Distributed Computing; Wiley Series on Parallel and D
• Subject: Natural sciences, medicine, IT, technology > IT, data processing > IT