ROOT::TThreadExecutor Class Reference

This class provides a simple interface to execute the same task multiple times in parallel threads, possibly with different arguments every time.

ROOT::TThreadExecutor::Map

This class inherits its interfaces from ROOT::TExecutorCRTP
, adapting them for multithreaded parallelism and extends them supporting:

• Parallel Foreach operations.
• Custom task granularity and partial reduction, by specifying reduction function and the number of chunks as extra parameters for the Map call. This is specially useful to reduce the size of intermediate results when dealing with a sizeable number of elements in the input data.

The two possible usages of the Map method are:

• Map(F func, unsigned nTimes): func is executed nTimes with no arguments
• Map(F func, T& args): func is executed on each element of the collection of arguments args

For either signature, func is executed as many times as needed by a pool of nThreads threads, where nThreads typically defaults to the number of cores.
A collection containing the result of each execution is returned.
Note: the user is responsible for the deletion of any object that might be created upon execution of func, returned objects included: ROOT::TThreadExecutor never deletes what it returns, it simply forgets it.

Parameters

func	a callable object, such as a lambda expression, an std::function, a functor object or a function that takes zero arguments (for the first signature) or one (for the second signature).
args	a standard vector, a ROOT::TSeq of integer type or an initializer list for the second signature. An integer only for the first.

Note: in cases where the function to be executed takes more than zero/one argument but all are fixed except zero/one, the function can be wrapped in a lambda or via std::bind to give it the right signature.

Return value:

An std::vector. The elements in the container will be the objects returned by func.

Examples:

root[] ROOT::TThreadExecutor pool; auto hists = pool.Map(CreateHisto, 10);
root[] ROOT::TThreadExecutor pool(2); auto squares = pool.Map([](int a) { return a*a; }, {1,2,3});

ROOT::TThreadExecutor::MapReduce

This set of methods behaves exactly like Map, but takes an additional function as a third argument. This function is applied to the set of objects returned by the corresponding Map execution to "squash" them into a single object. This function should be independent of the size of the vector returned by Map due to optimization of the number of chunks.

If this function is a binary operator, the "squashing" will be performed in parallel. This is exclusive to ROOT::TThreadExecutor and not any other ROOT::TExecutorCRTP-derived classes.
An integer can be passed as the fourth argument indicating the number of chunks we want to divide our work in. This may be useful to avoid the overhead introduced when running really short tasks.

Examples:

root[] ROOT::TThreadExecutor pool; auto ten = pool.MapReduce([]() { return 1; }, 10, [](const std::vector<int> &v) { return std::accumulate(v.begin(), v.end(), 0); })
root[] ROOT::TThreadExecutor pool; auto hist = pool.MapReduce(CreateAndFillHists, 10, PoolUtils::ReduceObjects);

Definition at line 41 of file TThreadExecutor.hxx.

Public Member Functions
	TThreadExecutor (const TThreadExecutor &)=delete
	TThreadExecutor (UInt_t nThreads=0u)
	Class constructor.
template<class F , class T >
void	Foreach (F func, const std::vector< T > &args, unsigned nChunks=0)
	Execute a function in parallel over the elements of a immutable vector, dividing the execution in nChunks.
template<class F , class INTEGER >
void	Foreach (F func, ROOT::TSeq< INTEGER > args, unsigned nChunks=0)
	Execute a function in parallel over a sequence of indexes, dividing the execution in nChunks.
template<class F , class T >
void	Foreach (F func, std::initializer_list< T > args, unsigned nChunks=0)
	Execute a function in parallel over the elements of an initializer_list, dividing the execution in nChunks.
template<class F , class T >
void	Foreach (F func, std::vector< T > &args, unsigned nChunks=0)
	Execute a function in parallel over the elements of a vector, dividing the execution in nChunks.
template<class F >
void	Foreach (F func, unsigned nTimes, unsigned nChunks=0)
	Execute a function without arguments several times in parallel, dividing the execution in nChunks.
unsigned	GetPoolSize () const
	Returns the number of worker threads in the task arena.
template<class F , class T , class R , class Cond = validMapReturnCond<F, T>>
auto	MapReduce (F func, const std::vector< T > &args, R redfunc) -> InvokeResult_t< F, T >
	Execute a function over the elements of an immutable vector in parallel (Map) and accumulate the results into a single value (Reduce).
template<class F , class T , class R , class Cond = validMapReturnCond<F, T>>
auto	MapReduce (F func, const std::vector< T > &args, R redfunc, unsigned nChunks) -> InvokeResult_t< F, T >
	Execute a function in parallel over the elements of an immutable vector (Map) and accumulate the results into a single value (Reduce).
template<class F , class INTEGER , class R , class Cond = validMapReturnCond<F, INTEGER>>
auto	MapReduce (F func, ROOT::TSeq< INTEGER > args, R redfunc, unsigned nChunks) -> InvokeResult_t< F, INTEGER >
	Execute a function in parallel over the elements of a vector (Map) and accumulate the results into a single value (Reduce).
template<class F , class T , class R , class Cond = validMapReturnCond<F, T>>
auto	MapReduce (F func, std::initializer_list< T > args, R redfunc, unsigned nChunks) -> InvokeResult_t< F, T >
	Execute a function in parallel over the elements of an initializer_list (Map) and accumulate the results into a single value (Reduce).
template<class F , class T , class R , class Cond = validMapReturnCond<F, T>>
auto	MapReduce (F func, std::vector< T > &args, R redfunc) -> InvokeResult_t< F, T >
	Execute a function over the elements of a vector in parallel (Map) and accumulate the results into a single value (Reduce).
template<class F , class T , class R , class Cond = validMapReturnCond<F, T>>
auto	MapReduce (F func, std::vector< T > &args, R redfunc, unsigned nChunks) -> InvokeResult_t< F, T >
	Execute a function in parallel over the elements of a vector (Map) and accumulate the results into a single value (Reduce).
template<class F , class R , class Cond = validMapReturnCond<F>>
auto	MapReduce (F func, unsigned nTimes, R redfunc) -> InvokeResult_t< F >
	Execute a function nTimes in parallel (Map) and accumulate the results into a single value (Reduce).
template<class F , class R , class Cond = validMapReturnCond<F>>
auto	MapReduce (F func, unsigned nTimes, R redfunc, unsigned nChunks) -> InvokeResult_t< F >
	Execute a function in parallel over the elements of a vector (Map) and accumulate the results into a single value (Reduce).
TThreadExecutor &	operator= (const TThreadExecutor &)=delete
template<class T , class BINARYOP >
auto	Reduce (const std::vector< T > &objs, BINARYOP redfunc) -> decltype(redfunc(objs.front(), objs.front()))
	"Reduce" an std::vector into a single object in parallel by passing a binary function as the second argument defining the reduction operation.
template<class T , class R >
auto	Reduce (const std::vector< T > &objs, R redfunc) -> decltype(redfunc(objs))
	"Reduce" an std::vector into a single object by passing a function as the second argument defining the reduction operation.
Public Member Functions inherited from ROOT::TExecutorCRTP< TThreadExecutor >
	TExecutorCRTP ()=default
	TExecutorCRTP (const TExecutorCRTP &)=delete
auto	Map (F func, const std::vector< T > &args) -> std::vector< InvokeResult_t< F, T > >
	Execute a function over the elements of an immutable vector.
auto	Map (F func, ROOT::TSeq< INTEGER > args) -> std::vector< InvokeResult_t< F, INTEGER > >
	Execute a function over a sequence of indexes.
auto	Map (F func, std::initializer_list< T > args) -> std::vector< InvokeResult_t< F, T > >
	Execute a function over the elements of an initializer_list.
auto	Map (F func, std::vector< T > &args) -> std::vector< InvokeResult_t< F, T > >
	Execute a function over the elements of a vector.
auto	Map (F func, unsigned nTimes) -> std::vector< InvokeResult_t< F > >
	Execute a function without arguments several times.
T *	MapReduce (F func, const std::vector< T * > &args)
	Execute a function over the TObject-inheriting elements of an immutable vector (Map) and merge the objects into a single one (Reduce).
auto	MapReduce (F func, const std::vector< T > &args, R redfunc) -> InvokeResult_t< F, T >
	Execute a function over the elements of an immutable vector (Map) and accumulate the results into a single value (Reduce).
auto	MapReduce (F func, ROOT::TSeq< INTEGER > args, R redfunc) -> InvokeResult_t< F, INTEGER >
	Execute a function over a sequence of indexes (Map) and accumulate the results into a single value (Reduce).
auto	MapReduce (F func, std::initializer_list< T > args, R redfunc) -> InvokeResult_t< F, T >
	Execute a function over the elements of an initializer_list (Map) and accumulate the results into a single value (Reduce).
T *	MapReduce (F func, std::vector< T * > &args)
	Execute a function over the TObject-inheriting elements of a vector (Map) and merge the objects into a single one (Reduce).
auto	MapReduce (F func, std::vector< T > &args, R redfunc) -> InvokeResult_t< F, T >
	Execute a function over the elements of a vector (Map) and accumulate the results into a single value (Reduce).
auto	MapReduce (F func, unsigned nTimes, R redfunc) -> InvokeResult_t< F >
	Execute a function without arguments several times (Map) and accumulate the results into a single value (Reduce).
TExecutorCRTP &	operator= (const TExecutorCRTP &)=delete
T *	Reduce (const std::vector< T * > &mergeObjs)
	"Reduce" an std::vector into a single object by using the object's Merge method.
auto	Reduce (const std::vector< T > &objs, R redfunc) -> decltype(redfunc(objs))
	"Reduce" an std::vector into a single object by passing a function as the second argument defining the reduction operation.

Private Member Functions
template<class F , class T , class R , class Cond = validMapReturnCond<F, T>>
auto	Map (F func, const std::vector< T > &args, R redfunc, unsigned nChunks) -> std::vector< InvokeResult_t< F, T > >
	Execute a function in parallel over the elements of an immutable vector, dividing the execution in nChunks and providing a result per chunk.
template<class F , class INTEGER , class R , class Cond = validMapReturnCond<F, INTEGER>>
auto	Map (F func, ROOT::TSeq< INTEGER > args, R redfunc, unsigned nChunks) -> std::vector< InvokeResult_t< F, INTEGER > >
	Execute a function in parallel over the elements of a sequence, dividing the execution in nChunks and providing a result per chunk.
template<class F , class T , class R , class Cond = validMapReturnCond<F, T>>
auto	Map (F func, std::initializer_list< T > args, R redfunc, unsigned nChunks) -> std::vector< InvokeResult_t< F, T > >
	Execute a function in parallel over the elements of an initializer_list, dividing the execution in nChunks and providing a result per chunk.
template<class F , class T , class R , class Cond = validMapReturnCond<F, T>>
auto	Map (F func, std::vector< T > &args, R redfunc, unsigned nChunks) -> std::vector< InvokeResult_t< F, T > >
	Execute a function in parallel over the elements of a vector, dividing the execution in nChunks and providing a result per chunk.
template<class F , class R , class Cond = validMapReturnCond<F>>
auto	Map (F func, unsigned nTimes, R redfunc, unsigned nChunks) -> std::vector< InvokeResult_t< F > >
	Execute a function nTimes in parallel, dividing the execution in nChunks and providing a result per chunk.
template<class F , class T , class Cond = validMapReturnCond<F, T>>
auto	MapImpl (F func, const std::vector< T > &args) -> std::vector< InvokeResult_t< F, T > >
	Execute a function over the elements of a vector in parallel.
template<class F , class INTEGER , class Cond = validMapReturnCond<F, INTEGER>>
auto	MapImpl (F func, ROOT::TSeq< INTEGER > args) -> std::vector< InvokeResult_t< F, INTEGER > >
	Execute a function over a sequence of indexes in parallel.
template<class F , class T , class Cond = validMapReturnCond<F, T>>
auto	MapImpl (F func, std::vector< T > &args) -> std::vector< InvokeResult_t< F, T > >
	Execute a function over the elements of a vector in parallel.
template<class F , class Cond = validMapReturnCond<F>>
auto	MapImpl (F func, unsigned nTimes) -> std::vector< InvokeResult_t< F > >
	Execute a function without arguments several times in parallel.
void	ParallelFor (unsigned start, unsigned end, unsigned step, const std::function< void(unsigned int i)> &f)
	Execute a function in parallel over the indices of a loop.
double	ParallelReduce (const std::vector< double > &objs, const std::function< double(double a, double b)> &redfunc)
	"Reduce" in parallel an std::vector<double> into a single double value
float	ParallelReduce (const std::vector< float > &objs, const std::function< float(float a, float b)> &redfunc)
	"Reduce" in parallel an std::vector<float> into a single float value
template<class T , class R >
auto	SeqReduce (const std::vector< T > &objs, R redfunc) -> decltype(redfunc(objs))
	"Reduce", sequentially, an std::vector into a single object

Private Attributes
std::shared_ptr< ROOT::Internal::RTaskArenaWrapper >	fTaskArenaW = nullptr
	Pointer to the TBB task arena wrapper.
friend	TExecutorCRTP

Additional Inherited Members
Protected Types inherited from ROOT::TExecutorCRTP< TThreadExecutor >
using	InvokeResult_t = ROOT::TypeTraits::InvokeResult_t< F, Args... >
using	validMapReturnCond = std::enable_if_t<!std::is_reference< InvokeResult_t< F, T... > >::value &&!std::is_void< InvokeResult_t< F, T... > >::value >
	type definition used in templated functions for not allowing mapping functions that return references or void.

#include <ROOT/TThreadExecutor.hxx>

Inheritance diagram for ROOT::TThreadExecutor:

Constructor & Destructor Documentation

TThreadExecutor() [1/2]

ROOT::TThreadExecutor::TThreadExecutor ( UInt_t  nThreads = 0u )

explicit

Class constructor.

If the scheduler is active (e.g. because another TThreadExecutor is in flight, or ROOT::EnableImplicitMT() was called), work with the current pool of threads. If not, initialize the pool of threads, spawning nThreads. nThreads' default value, 0, initializes the pool with as many logical threads as are available in the system (see NLogicalCores in RTaskArenaWrapper.cxx).

At construction time, TThreadExecutor automatically enables ROOT's thread-safety locks as per calling ROOT::EnableThreadSafety().

Definition at line 149 of file TThreadExecutor.cxx.

TThreadExecutor() [2/2]

ROOT::TThreadExecutor::TThreadExecutor ( const TThreadExecutor &  )

delete

Member Function Documentation

Foreach() [1/5]

template<class F , class T >

void ROOT::TThreadExecutor::Foreach	(	F	func,
		const std::vector< T > &	args,
		unsigned	nChunks = 0
	)

Execute a function in parallel over the elements of a immutable vector, dividing the execution in nChunks.

Parameters

func	Function to be executed on the elements of the vector passed as second parameter.
args	Immutable vector of elements passed as an argument to func.
nChunks	Number of chunks to split the input data for processing.

Definition at line 231 of file TThreadExecutor.hxx.

Foreach() [2/5]

template<class F , class INTEGER >

void ROOT::TThreadExecutor::Foreach	(	F	func,
		ROOT::TSeq< INTEGER >	args,
		unsigned	nChunks = 0
	)

Execute a function in parallel over a sequence of indexes, dividing the execution in nChunks.

Parameters

func	Function to be executed. Must take an element of the sequence passed assecond argument as a parameter.
args	Sequence of indexes to execute func on.
nChunks	Number of chunks to split the input data for processing.

Definition at line 169 of file TThreadExecutor.hxx.

Foreach() [3/5]

template<class F , class T >

void ROOT::TThreadExecutor::Foreach	(	F	func,
		std::initializer_list< T >	args,
		unsigned	nChunks = 0
	)

Execute a function in parallel over the elements of an initializer_list, dividing the execution in nChunks.

Parameters

func	Function to be executed on the elements of the initializer_list passed as second parameter.
args	initializer_list for a vector to apply func on.
nChunks	Number of chunks to split the input data for processing.

Definition at line 195 of file TThreadExecutor.hxx.

Foreach() [4/5]

template<class F , class T >

void ROOT::TThreadExecutor::Foreach	(	F	func,
		std::vector< T > &	args,
		unsigned	nChunks = 0
	)

Execute a function in parallel over the elements of a vector, dividing the execution in nChunks.

Parameters

func	Function to be executed on the elements of the vector passed as second parameter.
args	Vector of elements passed as an argument to func.
nChunks	Number of chunks to split the input data for processing.

Definition at line 207 of file TThreadExecutor.hxx.

Foreach() [5/5]

template<class F >

void ROOT::TThreadExecutor::Foreach	(	F	func,
		unsigned	nTimes,
		unsigned	nChunks = 0
	)

Execute a function without arguments several times in parallel, dividing the execution in nChunks.

Parameters

func	Function to be executed.
nTimes	Number of times function should be called.
nChunks	Number of chunks to split the input data for processing.

Definition at line 146 of file TThreadExecutor.hxx.

GetPoolSize()

unsigned ROOT::TThreadExecutor::GetPoolSize

(

)

const

Returns the number of worker threads in the task arena.

Returns

the number of worker threads assigned to the task arena.

Definition at line 216 of file TThreadExecutor.cxx.

Map() [1/5]

template<class F , class T , class R , class Cond >

auto ROOT::TThreadExecutor::Map ( F  func, const std::vector< T > &  args, R  redfunc, unsigned  nChunks  ) -> std::vector<InvokeResult_t<F, T>>

private

Execute a function in parallel over the elements of an immutable vector, dividing the execution in nChunks and providing a result per chunk.

Definition at line 436 of file TThreadExecutor.hxx.

Map() [2/5]

template<class F , class INTEGER , class R , class Cond >

auto ROOT::TThreadExecutor::Map ( F  func, ROOT::TSeq< INTEGER >  args, R  redfunc, unsigned  nChunks  ) -> std::vector<InvokeResult_t<F, INTEGER>>

private

Execute a function in parallel over the elements of a sequence, dividing the execution in nChunks and providing a result per chunk.

Definition at line 368 of file TThreadExecutor.hxx.

Map() [3/5]

template<class F , class T , class R , class Cond >

auto ROOT::TThreadExecutor::Map ( F  func, std::initializer_list< T >  args, R  redfunc, unsigned  nChunks  ) -> std::vector<InvokeResult_t<F, T>>

private

Execute a function in parallel over the elements of an initializer_list, dividing the execution in nChunks and providing a result per chunk.

Definition at line 470 of file TThreadExecutor.hxx.

Map() [4/5]

template<class F , class T , class R , class Cond >

auto ROOT::TThreadExecutor::Map ( F  func, std::vector< T > &  args, R  redfunc, unsigned  nChunks  ) -> std::vector<InvokeResult_t<F, T>>

private

Execute a function in parallel over the elements of a vector, dividing the execution in nChunks and providing a result per chunk.

Definition at line 402 of file TThreadExecutor.hxx.

Map() [5/5]

template<class F , class R , class Cond >

auto ROOT::TThreadExecutor::Map ( F  func, unsigned  nTimes, R  redfunc, unsigned  nChunks  ) -> std::vector<InvokeResult_t<F>>

private

Execute a function nTimes in parallel, dividing the execution in nChunks and providing a result per chunk.

Definition at line 289 of file TThreadExecutor.hxx.

MapImpl() [1/4]

template<class F , class T , class Cond >

auto ROOT::TThreadExecutor::MapImpl ( F  func, const std::vector< T > &  args  ) -> std::vector<InvokeResult_t<F, T>>

private

Execute a function over the elements of a vector in parallel.

Implementation of the Map method.

Parameters

func	Function to be executed on the elements of the vector passed as second parameter.
args	Vector of elements passed as an argument to func.

Returns

A vector with the results of the function calls.

Definition at line 344 of file TThreadExecutor.hxx.

MapImpl() [2/4]

template<class F , class INTEGER , class Cond >

auto ROOT::TThreadExecutor::MapImpl ( F  func, ROOT::TSeq< INTEGER >  args  ) -> std::vector<InvokeResult_t<F, INTEGER>>

private

Execute a function over a sequence of indexes in parallel.

Implementation of the Map method.

Parameters

func	Function to be executed. Must take an element of the sequence passed assecond argument as a parameter.
args	Sequence of indexes to execute func on.

Returns

A vector with the results of the function calls.

Definition at line 273 of file TThreadExecutor.hxx.

MapImpl() [3/4]

template<class F , class T , class Cond >

auto ROOT::TThreadExecutor::MapImpl ( F  func, std::vector< T > &  args  ) -> std::vector<InvokeResult_t<F, T>>

private

Execute a function over the elements of a vector in parallel.

Implementation of the Map method.

Parameters

func	Function to be executed on the elements of the vector passed as second parameter.
args	Vector of elements passed as an argument to func.

Returns

A vector with the results of the function calls.

Definition at line 320 of file TThreadExecutor.hxx.

MapImpl() [4/4]

template<class F , class Cond >

auto ROOT::TThreadExecutor::MapImpl ( F  func, unsigned  nTimes  ) -> std::vector<InvokeResult_t<F>>

private

Execute a function without arguments several times in parallel.

Implementation of the Map method.

Parameters

func	Function to be executed.
nTimes	Number of times function should be called.

Returns

A vector with the results of the function calls. Functions that take arguments can be executed (with fixed arguments) by wrapping them in a lambda or with std::bind.

Definition at line 254 of file TThreadExecutor.hxx.

MapReduce() [1/8]

template<class F , class T , class R , class Cond >

auto ROOT::TThreadExecutor::MapReduce	(	F	func,
		const std::vector< T > &	args,
		R	redfunc
	)		-> InvokeResult_t<F, T>

Execute a function over the elements of an immutable vector in parallel (Map) and accumulate the results into a single value (Reduce).

Parameters

func	Function to be executed on the elements of the vector passed as second parameter.
args	Immutable vector of elements passed as an argument to func.
redfunc	Reduction function to combine the results of the calls to func. Must return the same type as func.

Returns

A value result of "reducing" the vector returned by the Map operation into a single object.

Definition at line 535 of file TThreadExecutor.hxx.

MapReduce() [2/8]

template<class F , class T , class R , class Cond >

auto ROOT::TThreadExecutor::MapReduce	(	F	func,
		const std::vector< T > &	args,
		R	redfunc,
		unsigned	nChunks
	)		-> InvokeResult_t<F, T>

Execute a function in parallel over the elements of an immutable vector (Map) and accumulate the results into a single value (Reduce).

Benefits from partial reduction into nChunks intermediate results.

Benefits from partial reduction into nChunks intermediate results if the execution policy is multithreaded. Otherwise, it ignores the nChunks argument and performs a normal MapReduce operation.

Parameters

func	Function to be executed. Must take an element of the sequence passed assecond argument as a parameter.
args	Immutable vector, whose elements are passed as an argument to func.
redfunc	Reduction function to combine the results of the calls to func into partial results, and these into a final result. Must return the same type as func and should be callable with const std::vector<T> where T is the output of func.
nChunks	Number of chunks to split the input data for processing.

Returns

A value result of "reducing" the vector returned by the Map operation into a single object.

Definition at line 557 of file TThreadExecutor.hxx.

MapReduce() [3/8]

template<class F , class INTEGER , class R , class Cond >

auto ROOT::TThreadExecutor::MapReduce	(	F	func,
		ROOT::TSeq< INTEGER >	args,
		R	redfunc,
		unsigned	nChunks
	)		-> InvokeResult_t<F, INTEGER>

Execute a function in parallel over the elements of a vector (Map) and accumulate the results into a single value (Reduce).

Benefits from partial reduction into nChunks intermediate results.

Benefits from partial reduction into nChunks intermediate results if the execution policy is multithreaded. Otherwise, it ignores the nChunks argument and performs a normal MapReduce operation.

Parameters

func	Function to be executed. Must take an element of the sequence passed assecond argument as a parameter.
args	Sequence of indexes to execute func on.
redfunc	Reduction function to combine the results of the calls to func into partial results, and these into a final result. Must return the same type as func and should be callable with std::vector<T> where T is the output of func.
nChunks	Number of chunks to split the input data for processing.

Returns

A value result of "reducing" the vector returned by the Map operation into a single object.

Definition at line 504 of file TThreadExecutor.hxx.

MapReduce() [4/8]

template<class F , class T , class R , class Cond >

auto ROOT::TThreadExecutor::MapReduce	(	F	func,
		std::initializer_list< T >	args,
		R	redfunc,
		unsigned	nChunks
	)		-> InvokeResult_t<F, T>

Execute a function in parallel over the elements of an initializer_list (Map) and accumulate the results into a single value (Reduce).

Benefits from partial reduction into nChunks intermediate results.

Benefits from partial reduction into nChunks intermediate results if the execution policy is multithreaded. Otherwise, it ignores the nChunks argument and performs a normal MapReduce operation.

Parameters

func	Function to be executed. Must take an element of the sequence passed as second argument as a parameter.
args	initializer_list for a vector to apply func on.
redfunc	Reduction function to combine the results of the calls to func into partial results, and these into a final result. Must return the same type as func and should be callable with const std::vector<T> where T is the output of func.
nChunks	Number of chunks to split the input data for processing.

Returns

A value result of "reducing" the vector returned by the Map operation into a single object.

Definition at line 516 of file TThreadExecutor.hxx.

MapReduce() [5/8]

template<class F , class T , class R , class Cond >

auto ROOT::TThreadExecutor::MapReduce	(	F	func,
		std::vector< T > &	args,
		R	redfunc
	)		-> InvokeResult_t<F, T>

Execute a function over the elements of a vector in parallel (Map) and accumulate the results into a single value (Reduce).

Parameters

func	Function to be executed on the elements of the vector passed as second parameter.
args	Vector of elements passed as an argument to func.
redfunc	Reduction function to combine the results of the calls to func. Must return the same type as func.

Returns

A value result of "reducing" the vector returned by the Map operation into a single object.

Definition at line 526 of file TThreadExecutor.hxx.

MapReduce() [6/8]

template<class F , class T , class R , class Cond >

auto ROOT::TThreadExecutor::MapReduce	(	F	func,
		std::vector< T > &	args,
		R	redfunc,
		unsigned	nChunks
	)		-> InvokeResult_t<F, T>

Execute a function in parallel over the elements of a vector (Map) and accumulate the results into a single value (Reduce).

Benefits from partial reduction into nChunks intermediate results.

Benefits from partial reduction into nChunks intermediate results if the execution policy is multithreaded. Otherwise, it ignores the nChunks argument and performs a normal MapReduce operation.

Parameters

func	Function to be executed. Must take an element of the sequence passed assecond argument as a parameter.
args	Vector of elements passed as an argument to func.
redfunc	Reduction function to combine the results of the calls to func into partial results, and these into a final result. Must return the same type as func and should be callable with const std::vector<T> where T is the output of func.
nChunks	Number of chunks to split the input data for processing.

Returns

A value result of "reducing" the vector returned by the Map operation into a single object.

Definition at line 546 of file TThreadExecutor.hxx.

MapReduce() [7/8]

template<class F , class R , class Cond >

auto ROOT::TThreadExecutor::MapReduce	(	F	func,
		unsigned	nTimes,
		R	redfunc
	)		-> InvokeResult_t<F>

Execute a function nTimes in parallel (Map) and accumulate the results into a single value (Reduce).

Parameters

func	Function to be executed.
nTimes	Number of times function should be called.

Returns

A vector with the results of the function calls.

Parameters

redfunc

Reduction function to combine the results of the calls to func. Must return the same type as func.

Returns

A value result of "reducing" the vector returned by the Map operation into a single object.

Definition at line 482 of file TThreadExecutor.hxx.

MapReduce() [8/8]

template<class F , class R , class Cond >

auto ROOT::TThreadExecutor::MapReduce	(	F	func,
		unsigned	nTimes,
		R	redfunc,
		unsigned	nChunks
	)		-> InvokeResult_t<F>

Execute a function in parallel over the elements of a vector (Map) and accumulate the results into a single value (Reduce).

Benefits from partial reduction into nChunks intermediate results.

Benefits from partial reduction into nChunks intermediate results if the execution policy is multithreaded. Otherwise, it ignores the nChunks argument and performs a normal MapReduce operation.

Parameters

func	Function to be executed. Must take an element of the sequence passed as second argument as a parameter.
nTimes	Number of times function should be called.
redfunc	Reduction function to combine the results of the calls to func into partial results, and these into a final result. Must return the same type as func and should be callable with const std::vector<T> where T is the output of func.
nChunks	Number of chunks to split the input data for processing.

Returns

A value result of "reducing" the vector returned by the Map operation into a single object.

Definition at line 493 of file TThreadExecutor.hxx.

operator=()

TThreadExecutor & ROOT::TThreadExecutor::operator= ( const TThreadExecutor &  )

delete

ParallelFor()

void ROOT::TThreadExecutor::ParallelFor ( unsigned  start, unsigned  end, unsigned  step, const std::function< void(unsigned int i)> &  f  )

private

Execute a function in parallel over the indices of a loop.

Parameters

start	Start index of the loop.
end	End index of the loop.
step	Step size of the loop.
f	function to execute.

Definition at line 161 of file TThreadExecutor.cxx.

ParallelReduce() [1/2]

double ROOT::TThreadExecutor::ParallelReduce ( const std::vector< double > &  objs, const std::function< double(double a, double b)> &  redfunc  )

private

"Reduce" in parallel an std::vector<double> into a single double value

Parameters

objs	A vector of elements to combine.
redfunc	Reduction function to combine the elements of the vector objs.

Returns

A value result of combining the vector elements into a single object of the same type.

Definition at line 183 of file TThreadExecutor.cxx.

ParallelReduce() [2/2]

float ROOT::TThreadExecutor::ParallelReduce ( const std::vector< float > &  objs, const std::function< float(float a, float b)> &  redfunc  )

private

"Reduce" in parallel an std::vector<float> into a single float value

Parameters

objs	A vector of elements to combine.
redfunc	Reduction function to combine the elements of the vector objs.

Returns

A value result of combining the vector elements into a single object of the same type.

Definition at line 201 of file TThreadExecutor.cxx.

Reduce() [1/2]

template<class T , class BINARYOP >

auto ROOT::TThreadExecutor::Reduce	(	const std::vector< T > &	objs,
		BINARYOP	redfunc
	)		-> decltype(redfunc(objs.front(), objs.front()))

"Reduce" an std::vector into a single object in parallel by passing a binary function as the second argument defining the reduction operation.

Parameters

objs	A vector of elements to combine.
redfunc	Binary reduction function to combine the elements of the vector objs.

Returns

A value result of combining the vector elements into a single object of the same type.

Definition at line 581 of file TThreadExecutor.hxx.

Reduce() [2/2]

template<class T , class R >

auto ROOT::TThreadExecutor::Reduce	(	const std::vector< T > &	objs,
		R	redfunc
	)		-> decltype(redfunc(objs))

"Reduce" an std::vector into a single object by passing a function as the second argument defining the reduction operation.

Parameters

objs	A vector of elements to combine.
redfunc	Reduction function to combine the elements of the vector objs

Returns

A value result of combining the vector elements into a single object of the same type.

Definition at line 566 of file TThreadExecutor.hxx.

SeqReduce()

template<class T , class R >

auto ROOT::TThreadExecutor::SeqReduce ( const std::vector< T > &  objs, R  redfunc  ) -> decltype(redfunc(objs))

private

"Reduce", sequentially, an std::vector into a single object

Parameters

objs	A vector of elements to combine.
redfunc	Reduction function to combine the elements of the vector objs.

Returns

A value result of combining the vector elements into a single object of the same type.

Definition at line 595 of file TThreadExecutor.hxx.

Member Data Documentation

fTaskArenaW

std::shared_ptr<ROOT::Internal::RTaskArenaWrapper> ROOT::TThreadExecutor::fTaskArenaW = nullptr

private

Pointer to the TBB task arena wrapper.

Definition at line 134 of file TThreadExecutor.hxx.

TExecutorCRTP

friend ROOT::TThreadExecutor::TExecutorCRTP

private

Definition at line 42 of file TThreadExecutor.hxx.

Libraries for ROOT::TThreadExecutor:

---

The documentation for this class was generated from the following files:

• core/imt/inc/ROOT/TThreadExecutor.hxx
• core/imt/src/TThreadExecutor.cxx