Pro TBB : : C++ Parallel Programming with Threading Building Blocks.

• Author: Voss, Michael.
• TeilnehmendeR: Asenjo, Rafael.; Reinders, James.
• Place / Publishing House: Berkeley, CA : : Apress L. P.,, 2019.; ©2019.
• Year of Publication: 2019
• Edition: 1st ed.
• Language: English
• Physical Description: 1 online resource (807 pages)

Table of Contents

Intro -- Table of Contents -- About the Authors -- Acknowledgments -- Preface -- Part 1 -- Chapter 1: Jumping Right In: "Hello, TBB!" -- Why Threading Building Blocks? -- Performance: Small Overhead, Big Benefits for C++ -- Evolving Support for Parallelism in TBB and C++ -- Recent C++ Additions for Parallelism -- The Threading Building Blocks (TBB) Library -- Parallel Execution Interfaces -- Interfaces That Are Independent of the Execution Model -- Using the Building Blocks in TBB -- Let's Get Started Already! -- Getting the Threading Building Blocks (TBB) Library -- Getting a Copy of the Examples -- Writing a First "Hello, TBB!" Example -- Building the Simple Examples -- Steps to Set Up an Environment -- Building on Windows Using Microsoft Visual Studio -- Building on a Linux Platform from a Terminal -- Using the Intel Compiler -- tbbvars and pstlvars Scripts -- Setting Up Variables Manually Without Using the tbbvars Script or the Intel Compiler -- A More Complete Example -- Starting with a Serial Implementation -- Adding a Message-Driven Layer Using a Flow Graph -- Adding a Fork-Join Layer Using a parallel_for -- Adding a SIMD Layer Using a Parallel STL Transform -- Summary -- Chapter 2: Generic Parallel Algorithms -- Functional / Task Parallelism -- A Slightly More Complicated Example: A Parallel Implementation of Quicksort -- Loops: parallel_for, parallel_reduce, and parallel_scan -- parallel_for: Applying a Body to Each Element in a Range -- A Slightly More Complicated Example: Parallel Matrix Multiplication -- parallel_reduce: Calculating a Single Result Across a Range -- A Slightly More Complicated Example: Calculating π by Numerical Integration -- parallel_scan: A Reduction with Intermediate Values -- How Does This Work? -- A Slightly More Complicated Example: Line of Sight -- Cook Until Done: parallel_do and parallel_pipeline.

parallel_do: Apply a Body Until There Are No More Items Left -- A Slightly More Complicated Example: Forward Substitution -- parallel_pipeline: Streaming Items Through a Series of Filters -- A Slightly More Complicated Example: Creating 3D Stereoscopic Images -- Summary -- For More Information -- Chapter 3: Flow Graphs -- Why Use Graphs to Express Parallelism? -- The Basics of the TBB Flow Graph Interface -- Step 1: Create the Graph Object -- Step 2: Make the Nodes -- Step 3: Add Edges -- Step 4: Start the Graph -- Step 5: Wait for the Graph to Complete Executing -- A More Complicated Example of a Data Flow Graph -- Implementing the Example as a TBB Flow Graph -- Understanding the Performance of a Data Flow Graph -- The Special Case of Dependency Graphs -- Implementing a Dependency Graph -- Estimating the Scalability of a Dependency Graph -- Advanced Topics in TBB Flow Graphs -- Summary -- Chapter 4: TBB and the Parallel Algorithms of the C++ Standard Template Library -- Does the C++ STL Library Belong in This Book? -- A Parallel STL Execution Policy Analogy -- A Simple Example Using std::for_each -- What Algorithms Are Provided in a Parallel STL Implementation? -- How to Get and Use a Copy of Parallel STL That Uses TBB -- Algorithms in Intel's Parallel STL -- Capturing More Use Cases with Custom Iterators -- Highlighting Some of the Most Useful Algorithms -- std::for_each, std::for_each_n -- std::transform -- std::reduce -- std::transform_reduce -- A Deeper Dive into the Execution Policies -- The sequenced_policy -- The parallel_policy -- The unsequenced_policy -- The parallel_unsequenced_policy -- Which Execution Policy Should We Use? -- Other Ways to Introduce SIMD Parallelism -- Summary -- For More Information -- Chapter 5: Synchronization: Why and How to Avoid It -- A Running Example: Histogram of an Image -- An Unsafe Parallel Implementation.

A First Safe Parallel Implementation: Coarse-Grained Locking -- Mutex Flavors -- A Second Safe Parallel Implementation: Fine-Grained Locking -- A Third Safe Parallel Implementation: Atomics -- A Better Parallel Implementation: Privatization and Reduction -- Thread Local Storage, TLS -- enumerable_thread_specific, ETS -- combinable -- The Easiest Parallel Implementation: Reduction Template -- Recap of Our Options -- Summary -- For More Information -- Chapter 6: Data Structures for Concurrency -- Key Data Structures Basics -- Unordered Associative Containers -- Map vs. Set -- Multiple Values -- Hashing -- Unordered -- Concurrent Containers -- Concurrent Unordered Associative Containers -- concurrent_hash_map -- Concurrent Support for map/multimap and set/multiset Interfaces -- Built-In Locking vs. No Visible Locking -- Iterating Through These Structures Is Asking for Trouble -- Concurrent Queues: Regular, Bounded, and Priority -- Bounding Size -- Priority Ordering -- Staying Thread-Safe: Try to Forget About Top, Size, Empty, Front, Back -- Iterators -- Why to Use This Concurrent Queue: The A-B-A Problem -- When to NOT Use Queues: Think Algorithms! -- Concurrent Vector -- When to Use tbb::concurrent_vector Instead of std::vector -- Elements Never Move -- Concurrent Growth of concurrent_vectors -- Summary -- Chapter 7: Scalable Memory Allocation -- Modern C++ Memory Allocation -- Scalable Memory Allocation: What -- Scalable Memory Allocation: Why -- Avoiding False Sharing with Padding -- Scalable Memory Allocation Alternatives: Which -- Compilation Considerations -- Most Popular Usage (C/C++ Proxy Library): How -- Linux: malloc/new Proxy Library Usage -- macOS: malloc/new Proxy Library Usage -- Windows: malloc/new Proxy Library Usage -- Testing our Proxy Library Usage -- C Functions: Scalable Memory Allocators for C.

C++ Classes: Scalable Memory Allocators for C++ -- Allocators with std::allocator&lt -- T&gt -- Signature -- scalable_allocator -- tbb_allocator -- zero_allocator -- cached_aligned_allocator -- Memory Pool Support: memory_pool_allocator -- Array Allocation Support: aligned_space -- Replacing new and delete Selectively -- Performance Tuning: Some Control Knobs -- What Are Huge Pages? -- TBB Support for Huge Pages -- scalable_allocation_mode(int mode, intptr_t value) -- TBBMALLOC_USE_HUGE_PAGES -- TBBMALLOC_SET_SOFT_HEAP_LIMIT -- int scalable_allocation_command(int cmd, void ∗param) -- TBBMALLOC_CLEAN_ALL_BUFFERS -- TBBMALLOC_CLEAN_THREAD_BUFFERS -- Summary -- Chapter 8: Mapping Parallel Patterns to TBB -- Parallel Patterns vs. Parallel Algorithms -- Patterns Categorize Algorithms, Designs, etc. -- Patterns That Work -- Data Parallelism Wins -- Nesting Pattern -- Map Pattern -- Workpile Pattern -- Reduction Patterns (Reduce and Scan) -- Fork-Join Pattern -- Divide-and-Conquer Pattern -- Branch-and-Bound Pattern -- Pipeline Pattern -- Event-Based Coordination Pattern (Reactive Streams) -- Summary -- For More Information -- Part 2 -- Chapter 9: The Pillars of Composability -- What Is Composability? -- Nested Composition -- Concurrent Composition -- Serial Composition -- The Features That Make TBB a Composable Library -- The TBB Thread Pool (the Market) and Task Arenas -- The TBB Task Dispatcher: Work Stealing and More -- Putting It All Together -- Looking Forward -- Controlling the Number of Threads -- Work Isolation -- Task-to-Thread and Thread-to-Core Affinity -- Task Priorities -- Summary -- For More Information -- Chapter 10: Using Tasks to Create Your Own Algorithms -- A Running Example: The Sequence -- The High-Level Approach: parallel_invoke -- The Highest Among the Lower: task_group -- The Low-Level Task Interface: Part One - Task Blocking.

The Low-Level Task Interface: Part Two - Task Continuation -- Bypassing the Scheduler -- The Low-Level Task Interface: Part Three - Task Recycling -- Task Interface Checklist -- One More Thing: FIFO (aka Fire-and-Forget) Tasks -- Putting These Low-Level Features to Work -- Summary -- For More Information -- Chapter 11: Controlling the Number of Threads Used for Execution -- A Brief Recap of the TBB Scheduler Architecture -- Interfaces for Controlling the Number of Threads -- Controlling Thread Count with task_scheduler_init -- Controlling Thread Count with task_arena -- Controlling Thread Count with global_control -- Summary of Concepts and Classes -- The Best Approaches for Setting the Number of Threads -- Using a Single task_scheduler_init Object for a Simple Application -- Using More Than One task_scheduler_init Object in a Simple Application -- Using Multiple Arenas with Different Numbers of Slots to Influence Where TBB Places Its Worker Threads -- Using global_control to Control How Many Threads Are Available to Fill Arena Slots -- Using global_control to Temporarily Restrict the Number of Available Threads -- When NOT to Control the Number of Threads -- Figuring Out What's Gone Wrong -- Summary -- Chapter 12: Using Work Isolation for Correctness and Performance -- Work Isolation for Correctness -- Creating an Isolated Region with this_task_arena::isolate -- Oh No! Work Isolation Can Cause Its Own Correctness Issues! -- Even When It Is Safe, Work Isolation Is Not Free -- Using Task Arenas for Isolation: A Double-Edged Sword -- Don't Be Tempted to Use task_arenas to Create Work Isolation for Correctness -- Summary -- For More Information -- Chapter 13: Creating Thread-to-Core and Task-to-Thread Affinity -- Creating Thread-to-Core Affinity -- Creating Task-to-Thread Affinity -- When and How Should We Use the TBB Affinity Features? -- Summary.

For More Information.