Programming Massively Parallel Processors, 3rd Edition

Index

Note: Page numbers followed by “b”, “f” and “t” refer to boxes, figures, and tables, respectively.

A

Abrupt underflow convention, 137

Abstraction, 484–486

memory bandwidth of filling kernels, 485t

programmer productivity, 484

real-world performance, 485–486

robustness, 484

Accelerator, 517

device, 414

management, 525–528, 526f

Accelerator, class, 525–526

Accelerator::set_default static method, 526

Accelerator_view, C++ AMP, 526

AddVecKernel function, 36, 120

Adjacency matrix, 258

representation of simple graph, 259f

sparse matrix representation of, 259f

Adjacent synchronization, 193

Algorithm, 153–156

considerations, 140–142

selection, 374–379

ALU, See Arithmetic and logic unit (ALU)

AMD Opteron family, 1

Amdahl’s law, 10, 373–374, 374

ANSI C code, ...