Get PDF File in ./pdf
- 1.1 并行计算的动机
- 1.2 并行计算的应用
- 1.3 本书的组织结构
- 2.1 隐式并行 - 微处理器架构的趋势
- 2.2 内存系统性能的限制
- 2.3 并行计算平台的逻辑架构
- 2.4 并行平台的物理架构
- 2.5 并行计算中的通信开销
- 2.6 互联网络的路由机制
- 2.7 进程到处理器的映射技术及其影响
- 4.1 One-to-All Broadcast和All-to-One Reduction
- 4.2 All-to-All Broadcast和All-to-All Reduction
- 4.3 All-Reduce和Prefix-Sum操作
- 4.4 Scatter和Gather操作
- 4.5 All-to-All Personalized Communication
- 4.6 循环移位
- 4.7 提高通信速度的方法
- 4.8 总结
- 5.1 Sources of Overhead in Parallel Programs
- 5.2 Performance Metrics for Parallel Systems
- 5.3 The Effect of Granularity on Performance
- 5.4 Scalability of Parallel Systems
- 5.5 Minimum Execution Time and Minimum Cost-Optimal Execution Time
- 5.6 Asymptotic Analysis of Parallel Programs
- 5.7 Other Scalability Metrics
- 6.1 Principles of Message-Passing Programming
- 6.2 The Building Blocks: Send and Receive Operations
- 6.3 MPI: the Message Passing Interface
- 6.4 Topologies and Embedding
- 6.5 Overlapping Communication with Computation
- 6.6 Collective Communication and Computation Operations
- 6.7 Groups and Communicators
- 7.1 Thread Basics
- 7.2 Why Threads?
- 7.3 The POSIX Thread API
- 7.4 Thread Basics: Creation and Termination
- 7.5 Synchronization Primitives in Pthreads
- 7.6 Controlling Thread and Synchronization Attributes
- 7.7 Thread Cancellation
- 7.8 Composite Synchronization Constructs
- 7.9 Tips for Designing Asynchronous Programs
- 7.10 OpenMP: a Standard for Directive Based Parallel Programming
- 8.1 Matrix-Vector Multiplication
- 8.2 Matrix-Matrix Multiplication
- 8.3 Solving a System of Linear Equations
- 9.1 Issues in Sorting on Parallel Computers
- 9.2 Sorting Networks
- 9.3 Bubble Sort and its Variants
- 9.4 Quicksort
- 9.5 Bucket and Sample Sort
- 9.6 Other Sorting Algorithms
- 10.1 Definitions and Representation
- 10.2 Minimum Spanning Tree: Prim's Algorithm
- 10.3 Single-Source Shortest Paths: Dijkstra's Algorithm
- 10.4 All-Pairs Shortest Paths
- 10.5 Transitive Closure
- 10.6 Connected Components
- 10.7 Algorithms for Sparse Graphs
- 11.1 Definitions and Examples
- 11.2 Sequential Search Algorithms
- 11.3 Search Overhead Factor
- 11.4 Parallel Depth-First Search
- 11.5 Parallel Best-First Search
- 11.6 Speedup Anomalies in Parallel Search Algorithms