UPC – Distributed Shared Memory Programming
Distributed Shared Memory Programming
Samenvatting
This is the first book to explain the language Unified Parallel C and its use. Authors El–Ghazawi, Carlson, and Sterling are among the developers of UPC, with close links with the industrial members of the UPC consortium. Their text covers background material on parallel architectures and algorithms, and includes UPC programming case studies. This book represents an invaluable resource for the growing number of UPC users and applications developers. More information about UPC can be found at: http://upc.gwu.edu/
An Instructor Support FTP site is available from the Wiley editorial department.
Specificaties
Inhoudsopgave
<p>1. Introductory Tutorial 1</p>
<p>1.1 Getting Started 1</p>
<p>1.2 Private and Shared Data 3</p>
<p>1.3 Shared Arrays and Affinity of Shared Data 6</p>
<p>1.4 Synchronization and Memory Consistency 8</p>
<p>1.5 Work Sharing 10</p>
<p>1.6 UPC Pointers 11</p>
<p>1.7 Summary 14</p>
<p>Exercises 14</p>
<p>2. Programming View and UPC Data Types 17</p>
<p>2.1 Programming Models 17</p>
<p>2.2 UPC Programming Model 20</p>
<p>2.3 Shared and Private Variables 21</p>
<p>2.4 Shared and Private Arrays 23</p>
<p>2.5 Blocked Shared Arrays 25</p>
<p>2.6 Compiling Environments and Shared Arrays 30</p>
<p>2.7 Summary 30</p>
<p>Exercises 31</p>
<p>3. Pointers and Arrays 33</p>
<p>3.1 UPC Pointers 33</p>
<p>3.2 Pointer Arithmetic 35</p>
<p>3.3 Pointer Casting and Usage Practices 38</p>
<p>3.4 Pointer Information and Manipulation Functions 40</p>
<p>3.5 More Pointer Examples 43</p>
<p>3.6 Summary 47</p>
<p>Exercises 47</p>
<p>4. Work Sharing and Domain Decomposition 49</p>
<p>4.1 Basic Work Distribution 50</p>
<p>4.2 Parallel Iterations 51</p>
<p>4.3 Multidimensional Data 54</p>
<p>4.4 Distributing Trees 62</p>
<p>4.5 Summary 71</p>
<p>Exercises 71</p>
<p>5. Dynamic Shared Memory Allocation 73</p>
<p>5.1 Allocating a Global Shared Memory Space Collectively 73</p>
<p>5.2 Allocating Multiple Global Spaces 78</p>
<p>5.3 Allocating Local Shared Spaces 82</p>
<p>5.4 Freeing Allocated Spaces 89</p>
<p>5.5 Summary 90</p>
<p>Exercises 90</p>
<p>6. Synchronization and Memory Consistency 91</p>
<p>6.1 Barriers 92</p>
<p>6.2 Split–Phase Barriers 94</p>
<p>6.3 Locks 99</p>
<p>6.4 Memory Consistency 108</p>
<p>6.5 Summary 113</p>
<p>Exercises 114</p>
<p>7. Performance Tuning and Optimization 115</p>
<p>7.1 Parallel System Architectures 116</p>
<p>7.2 Performance Issues in Parallel Programming 120</p>
<p>7.3 Role of Compilers and Run–Time Systems 122</p>
<p>7.4 UPC Hand Optimization 123</p>
<p>7.5 Case Studies 128</p>
<p>7.6 Summary 135</p>
<p>Exercises 135</p>
<p>8. UPC Libraries 137</p>
<p>8.1 UPC Collective Library 137</p>
<p>8.2 UPC–IO Library 141</p>
<p>8.3 Summary 146</p>
<p>References 147</p>
<p>Appendix A: UPC Language Specifications, v1.1.1 149</p>
<p>Appendix B: UPC Collective Operations Specifications, v1.0 183</p>
<p>Appendix C: UPC–IO Specifications, v1.0 203</p>
<p>Appendix D: How to Compile and Run UPC Programs 243</p>
<p>Appendix E: Quick UPC Reference 245</p>
<p>Index 251</p>