Muutke küpsiste eelistusi

E-raamat: Applied Parallel Computing [World Scientific e-raamat]

(Stony Brook Univ, Usa)
  • Formaat: 220 pages
  • Ilmumisaeg: 11-Oct-2012
  • Kirjastus: World Scientific Publishing Co Pte Ltd
  • ISBN-13: 9789814307611
  • World Scientific e-raamat
  • Hind: 110,88 €*
  • * hind, mis tagab piiramatu üheaegsete kasutajate arvuga ligipääsu piiramatuks ajaks
  • Formaat: 220 pages
  • Ilmumisaeg: 11-Oct-2012
  • Kirjastus: World Scientific Publishing Co Pte Ltd
  • ISBN-13: 9789814307611
The book provides a practical guide to computational scientists and engineers to help advance their research by exploiting the superpower of supercomputers with many processors and complex networks. This book focuses on the design and analysis of basic parallel algorithms, the key components for composing larger packages for a wide range of applications.
Preface v
Chapter 1 Introduction
1(12)
1.1 Definition of Parallel Computing
1(3)
1.2 Evolution of Computers
4(4)
1.3 An Enabling Technology
8(1)
1.4 Cost Effectiveness
9(4)
Chapter 2 Performance Metrics and Models
13(6)
2.1 Parallel Activity Trace
13(1)
2.2 Speedup
14(1)
2.3 Parallel Efficiency
15(1)
2.4 Load Imbalance
15(1)
2.5 Granularity
16(1)
2.6 Overhead
17(1)
2.7 Scalability
18(1)
2.8 Amdahl's Law
18(1)
Chapter 3 Hardware Systems
19(16)
3.1 Node Architectures
19(2)
3.2 Network Interconnections
21(7)
3.3 Instruction and Data Streams
28(1)
3.4 Processor-Memory Connectivity
29(1)
3.5 IO Subsystems
29(2)
3.6 System Convergence
31(1)
3.7 Design Considerations
31(4)
Chapter 4 Software Systems
35(10)
4.1 Node Software
35(2)
4.2 Programming Models
37(6)
4.3 Parallel Debuggers
43(1)
4.4 Parallel Profilers
43(2)
Chapter 5 Design of Algorithms
45(20)
5.1 Algorithm Models
46(8)
5.2 Examples of Collective Operations
54(2)
5.3 Mapping Tasks to Processors
56(9)
Chapter 6 Linear Algebra
65(24)
6.1 Problem Decomposition
65(3)
6.2 Matrix Operations
68(13)
6.3 Solution of Linear Systems
81(8)
Chapter 7 Differential Equations
89(16)
7.1 Integration and Differentiation
89(3)
7.2 Partial Differential Equations
92(13)
Chapter 8 Fourier Transforms
105(10)
8.1 Fourier Transforms
105(1)
8.2 Discrete Fourier Transforms
106(1)
8.3 Fast Fourier Transforms
107(4)
8.4 Simple Parallelization
111(1)
8.5 The Transpose Method
112(1)
8.6 Complexity Analysis for FFT
113(2)
Chapter 9 Optimization
115(8)
9.1 Monte Carlo Methods
116(3)
9.2 Parallelization
119(4)
Chapter 10 Applications
123(16)
10.1 Newton's Equation and Molecular Dynamics
124(9)
10.2 Schrodinger's Equations and Quantum Mechanics
133(1)
10.3 Partition Function, DFT and Material Science
134(1)
10.4 Maxwell's Equations and Electrical Engineering
135(1)
10.5 Diffusion Equation and Mechanical Engineering
135(1)
10.6 Navier-Stokes Equation and CFD
136(1)
10.7 Other Applications
136(3)
Appendix A MPI
139(32)
A.1 An MPI Primer
139(20)
A.2 Examples of Using MPI
159(2)
A.3 MPI Tools
161(6)
A.4 Complete List of MPI Functions
167(4)
Appendix B OpenMP
171(10)
B.1 Introduction to OpenMP
171(1)
B.2 Memory Model of OpenMP
172(1)
B.3 OpenMP Directives
172(2)
B.4 Synchronization
174(1)
B.5 Runtime Library Routines
175(3)
B.6 Examples of Using OpenMP
178(2)
B.7 The Future
180(1)
Appendix C Projects
181(12)
Project C.1 Watts and Flops of Supercomputers
181(1)
Project C.2 Review of Supercomputers
181(1)
Project C.3 Top500 and BlueGene Supercomputers
181(1)
Project C.4 Say Hello in Order
182(1)
Project C.5 Broadcast on Torus
183(1)
Project C.6 Competing with MPI on Broadcast, Scatter, etc
183(1)
Project C.7 Simple Matrix Multiplication
183(1)
Project C.8 Matrix Multiplication on 4D Torus
183(1)
Project C.9 Matrix Multiplication and PAT
184(1)
Project C.10 Matrix Inversion
184(1)
Project C.11 Simple Analysis of an iBT Network
185(1)
Project C.12 Compute Eigenvalues of Adjacency Matrices of Networks
185(1)
Project C.13 Mapping Wave Equation to Torus
185(1)
Project C.14 Load Balance in 3D Mesh
186(1)
Project C.15 Wave Equation and PAT
186(1)
Project C.16 Computing Coulomb's Forces
187(1)
Project C.17 Timing Model for MD
187(1)
Project C.18 Minimizing Lennard-Jones Potential
188(1)
Project C.19 Install and Profile CP2K
188(1)
Project C.20 Install and Profile CPMD
189(1)
Project C.21 Install and Profile NAMD
190(1)
Project C.22 FFT on Beowulf
190(1)
Project C.23 FFT on BlueGene/Q
191(1)
Project C.24 Word Analysis
191(1)
Project C.25 Cost Estimate of a 0.1 Pflops System
191(1)
Project C.26 Design of a Pflops System
191(2)
Appendix D Program Examples
193(10)
D.1 Matrix-Vector Multiplication
193(2)
D.2 Long Range N-body Force
195(6)
D.3 Integration
201(2)
References 203(2)
Index 205