Visible to Intel only — GUID: GUID-DC707D6D-5DCA-4A2F-919F-C78E779CDD8E
Package Contents
Parallelizing Simple Loops
Parallelizing Complex Loops
Parallelizing Data Flow and Dependence Graphs
Work Isolation
Exceptions and Cancellation
Containers
Mutual Exclusion
Timing
Memory Allocation
The Task Scheduler
Design Patterns
Migrating from Threading Building Blocks (TBB)
Constrained APIs
Invoke a Callable Object
Appendix A Costs of Time Slicing
Appendix B Mixing With Other Threading Packages
References
parallel_for_each Body semantics and requirements
parallel_sort ranges interface extension
TBB_malloc_replacement_log Function
Type-specified message keys for join_node
Scalable Memory Pools
Helper Functions for Expressing Graphs
concurrent_lru_cache
task_group extensions
The customizing mutex type for concurrent_hash_map
Visible to Intel only — GUID: GUID-DC707D6D-5DCA-4A2F-919F-C78E779CDD8E
Design Patterns
This section provides some common parallel programming patterns and how to implement them in oneAPI Threading Building Blocks (oneTBB).
The description of each pattern has the following format:
Problem – describes the problem to be solved.
Context – describes contexts in which the problem arises.
Forces - considerations that drive use of the pattern.
Solution - describes how to implement the pattern.
Example – presents an example implementation.
Variations and examples are sometimes discussed. The code examples are intended to emphasize key points and are not full-fledged code. Examples may omit obvious const overloads of non-const methods.
Much of the nomenclature and examples are adapted from Web pages created by Eun-Gyu and Marc Snir, and the Berkeley parallel patterns wiki. See links in the General References section.
For brevity, some of the code examples use C++11 lambda expressions. It is straightforward, albeit sometimes tedious, to translate such lambda expressions into equivalent C++03 code.