template<typename C = DefaultClosureWrapper>

tf::DynamicPartitioner class

class to construct a dynamic partitioner for scheduling parallel algorithms

Template parameters
C	closure wrapper type (default tf::DefaultClosureWrapper)

The partitioner splits iterations into many partitions each of size equal to the given chunk size. Different partitions are distributed dynamically to workers without any specific order.

In addition to partition size, the application can specify a closure wrapper for a dynamic partitioner. A closure wrapper allows the application to wrapper a partitioned task (i.e., closure) with a custom function object that performs additional tasks. For example:

std::atomic<int> count = 0;
tf::Taskflow taskflow;
taskflow.for_each_index(0, 100, 1, 
  [](){                 
    printf("%d\n", i); 
  },
  tf::DynamicPartitioner(0, [](auto&& closure){
    // do something before invoking the partitioned task
    // ...
    
    // invoke the partitioned task
    closure();

    // do something else after invoking the partitioned task
    // ...
  }
);
executor.run(taskflow).wait();

Base classes

template<typename C = DefaultClosureWrapper>

class PartitionerBase<DefaultClosureWrapper>

class to derive a partitioner for scheduling parallel algorithms

Public static functions

static auto type() -> PartitionerType constexpr

queries the partition type (dynamic)

Constructors, destructors, conversion operators

DynamicPartitioner() defaulted

default constructor

DynamicPartitioner(size_t sz) explicit

construct a dynamic partitioner with the given chunk size

DynamicPartitioner(size_t sz, C&& closure) explicit

construct a dynamic partitioner with the given chunk size and the closure