Logo ROOT  
Reference Guide
 
Loading...
Searching...
No Matches
ROOT::TProcessExecutor Class Reference

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

ROOT::TProcessExecutor::Map

This class inherits its interfaces from ROOT::TExecutor
. 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 fNWorkers workers; the number of workers can be passed to the constructor or set via SetNWorkers. It 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::TProcessExecutor never deletes what it returns, it simply forgets it.
Note: that the usage of ROOT::TProcessExecutor::Map is indicated only when the task to be executed takes more than a few seconds, otherwise the overhead introduced by Map will outrun the benefits of parallel execution on most machines.

Parameters
funca 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).
argsa 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.
Note: the user should take care of initializing random seeds differently in each process (e.g. using the process id in the seed). Otherwise several parallel executions might generate the same sequence of pseudo-random numbers.

Return value:

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

Examples:

root[] ROOT::TProcessExecutor pool; auto hists = pool.Map(CreateHisto, 10);
root[] ROOT::TProcessExecutor pool(2); auto squares = pool.Map([](int a) { return a*a; }, {1,2,3});
#define a(i)
Definition RSha256.hxx:99
auto Map(F func, unsigned nTimes) -> std::vector< InvokeResult_t< F > >
Execute a function without arguments several times.
This class provides a simple interface to execute the same task multiple times in parallel,...

ROOT::TProcessExecutor::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.

Examples:

root[] ROOT::TProcessExecutor pool; auto ten = pool.MapReduce([]() { return 1; }, 10, [](const std::vector<int> &v) { return std::accumulate(v.begin(), v.end(), 0); })
root[] ROOT::TProcessExecutor pool; auto hist = pool.MapReduce(CreateAndFillHists, 10, PoolUtils::ReduceObjects);
Merge collection of TObjects.
Definition PoolUtils.h:35
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).

Definition at line 37 of file TProcessExecutor.hxx.

Public Member Functions

 TProcessExecutor (const TProcessExecutor &)=delete
 
 TProcessExecutor (unsigned nWorkers=0)
 Class constructor.
 
 ~TProcessExecutor ()=default
 
unsigned GetPoolSize () const
 Return the number of pooled parallel workers.
 
template<class F , class T , class R , class Cond = noReferenceCond<F, T>>
auto MapReduce (F func, const std::vector< T > &args, R redfunc) -> 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 T , class R , class Cond = noReferenceCond<F, T>>
auto MapReduce (F func, std::vector< T > &args, R redfunc) -> 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 = noReferenceCond<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).
 
TProcessExecutoroperator= (const TProcessExecutor &)=delete
 
void SetNWorkers (unsigned n)
 
- Public Member Functions inherited from ROOT::TExecutorCRTP< TProcessExecutor >
 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).
 
TExecutorCRTPoperator= (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 Types

enum class  ETask : unsigned char {
  kNoTask , kMap , kMapWithArg , kMapRed ,
  kMapRedWithArg
}
 A collection of the types of tasks that TProcessExecutor can execute. More...
 
template<typename F , typename... Args>
using InvokeResult_t = ROOT::TypeTraits::InvokeResult_t< F, Args... >
 

Private Member Functions

template<class T >
void Collect (std::vector< T > &reslist)
 Listen for messages sent by the workers and call the appropriate handler function.
 
template<class T >
void HandlePoolCode (MPCodeBufPair &msg, TSocket *sender, std::vector< T > &reslist)
 Handle message and reply to the worker.
 
template<class F , class T , class Cond = noReferenceCond<F, T>>
auto MapImpl (F func, const std::vector< T > &args) -> std::vector< InvokeResult_t< F, T > >
 Execute a function over the elements of an immutable vector in parallel Implementation of the Map method.
 
template<class F , class INTEGER , class Cond = noReferenceCond<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 = noReferenceCond<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 Implementation of the Map method.
 
template<class F , class Cond = noReferenceCond<F>>
auto MapImpl (F func, unsigned nTimes) -> std::vector< InvokeResult_t< F > >
 Execute a function without arguments several times in parallel.
 
void ReplyToFuncResult (TSocket *s)
 Reply to a worker who just sent a result.
 
void ReplyToIdle (TSocket *s)
 Reply to a worker who is idle.
 
void Reset ()
 Reset TProcessExecutor's state.
 
- Private Member Functions inherited from TMPClient
 TMPClient (const TMPClient &)=delete
 
 TMPClient (unsigned nWorkers=0)
 Class constructor.
 
 ~TMPClient ()
 Class destructor.
 
template<class T >
unsigned Broadcast (unsigned code, const std::vector< T > &objs)
 Send a message with a different object to each server.
 
template<class T >
unsigned Broadcast (unsigned code, std::initializer_list< T > &objs)
 Send a message with a different object to each server.
 
template<class T >
unsigned Broadcast (unsigned code, T obj, unsigned nMessages=0)
 Send a message containing code and obj to each worker, up to a maximum number of nMessages workers.
 
unsigned Broadcast (unsigned code, unsigned nMessages=0)
 Send a message with the specified code to at most nMessages workers.
 
void DeActivate (TSocket *s)
 DeActivate a certain socket.
 
bool Fork (TMPWorker &server)
 This method forks the ROOT session into fNWorkers children processes.
 
bool GetIsParent () const
 
TMonitorGetMonitor ()
 
unsigned GetNWorkers () const
 
void HandleMPCode (MPCodeBufPair &msg, TSocket *sender)
 Handle messages containing an EMPCode.
 
TMPClientoperator= (const TMPClient &)=delete
 
void ReapWorkers ()
 Wait on worker processes and remove their pids from fWorkerPids.
 
void Remove (TSocket *s)
 Remove a certain socket from the monitor.
 
void SetNWorkers (unsigned n)
 Set the number of workers that will be spawned by the next call to Fork()
 

Private Attributes

unsigned fNProcessed
 number of arguments already passed to the workers
 
unsigned fNToProcess
 total number of arguments to pass to the workers
 
ETask fTaskType = ETask::kNoTask
 the kind of task that is being executed, if any
 
friend TExecutorCRTP
 

Additional Inherited Members

- Public Types inherited from ROOT::TExecutorCRTP< TProcessExecutor >
using noReferenceCond = std::enable_if_t<"Function can't return a reference" &&!std::is_reference< InvokeResult_t< F, T... > >::value >
 type definition in used in templated functions for not allowing mapping functions that return references.
 

#include <ROOT/TProcessExecutor.hxx>

Inheritance diagram for ROOT::TProcessExecutor:
[legend]

Member Typedef Documentation

◆ InvokeResult_t

template<typename F , typename... Args>
using ROOT::TProcessExecutor::InvokeResult_t = ROOT::TypeTraits::InvokeResult_t<F, Args...>
private

Definition at line 41 of file TProcessExecutor.hxx.

Member Enumeration Documentation

◆ ETask

enum class ROOT::TProcessExecutor::ETask : unsigned char
strongprivate

A collection of the types of tasks that TProcessExecutor can execute.

It is used to interpret in the right way and properly reply to the messages received (see, for example, TProcessExecutor::HandleInput)

Enumerator
kNoTask 

no task is being executed

kMap 

a Map method with no arguments is being executed

kMapWithArg 

a Map method with arguments is being executed

kMapRed 

a MapReduce method with no arguments is being executed

kMapRedWithArg 

a MapReduce method with arguments is being executed

Definition at line 102 of file TProcessExecutor.hxx.

Constructor & Destructor Documentation

◆ TProcessExecutor() [1/2]

ROOT::TProcessExecutor::TProcessExecutor ( unsigned  nWorkers = 0)
explicit

Class constructor.

Parameters
nWorkersNumber of times this ROOT session will be forked, i.e. the number of workers that will be spawned.

Definition at line 90 of file TProcessExecutor.cxx.

◆ ~TProcessExecutor()

ROOT::TProcessExecutor::~TProcessExecutor ( )
default

◆ TProcessExecutor() [2/2]

ROOT::TProcessExecutor::TProcessExecutor ( const TProcessExecutor )
delete

Member Function Documentation

◆ Collect()

template<class T >
void ROOT::TProcessExecutor::Collect ( std::vector< T > &  reslist)
private

Listen for messages sent by the workers and call the appropriate handler function.

TProcessExecutor::HandlePoolCode is called on messages with a code < 1000 and TMPClient::HandleMPCode is called on messages with a code >= 1000.

Definition at line 415 of file TProcessExecutor.hxx.

◆ GetPoolSize()

unsigned ROOT::TProcessExecutor::GetPoolSize ( ) const
inline

Return the number of pooled parallel workers.

Returns
The number of workers in the pool.

Definition at line 75 of file TProcessExecutor.hxx.

◆ HandlePoolCode()

template<class T >
void ROOT::TProcessExecutor::HandlePoolCode ( MPCodeBufPair msg,
TSocket sender,
std::vector< T > &  reslist 
)
private

Handle message and reply to the worker.

Definition at line 386 of file TProcessExecutor.hxx.

◆ MapImpl() [1/4]

template<class F , class T , class Cond >
auto ROOT::TProcessExecutor::MapImpl ( F  func,
const std::vector< T > &  args 
) -> std::vector<InvokeResult_t<F, T>>
private

Execute a function over the elements of an immutable vector in parallel Implementation of the Map method.

Parameters
funcFunction to be executed on the elements of the vector passed as second parameter.
argsVector of elements passed as an argument to func.
Returns
A vector with the results of the function calls.

Definition at line 208 of file TProcessExecutor.hxx.

◆ MapImpl() [2/4]

template<class F , class INTEGER , class Cond >
auto ROOT::TProcessExecutor::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
funcFunction to be executed. Must take an element of the sequence passed assecond argument as a parameter.
argsSequence of indexes to execute func on.
Returns
A vector with the results of the function calls.

Definition at line 253 of file TProcessExecutor.hxx.

◆ MapImpl() [3/4]

template<class F , class T , class Cond >
auto ROOT::TProcessExecutor::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
funcFunction to be executed on the elements of the vector passed as second parameter.
argsVector of elements passed as an argument to func.
Returns
A vector with the results of the function calls.

Definition at line 163 of file TProcessExecutor.hxx.

◆ MapImpl() [4/4]

template<class F , class Cond >
auto ROOT::TProcessExecutor::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
funcFunction to be executed.
nTimesNumber 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 122 of file TProcessExecutor.hxx.

◆ MapReduce() [1/3]

template<class F , class T , class R , class Cond >
auto ROOT::TProcessExecutor::MapReduce ( F  func,
const std::vector< T > &  args,
R  redfunc 
) -> 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
funcFunction to be executed. Must take an element of the sequence passed assecond argument as a parameter.
argsImmutable vector, whose elements are passed as an argument to func.
redfuncReduction 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.
nChunksNumber 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 347 of file TProcessExecutor.hxx.

◆ MapReduce() [2/3]

template<class F , class T , class R , class Cond >
auto ROOT::TProcessExecutor::MapReduce ( F  func,
std::vector< T > &  args,
R  redfunc 
) -> 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
funcFunction to be executed. Must take an element of the sequence passed assecond argument as a parameter.
argsVector of elements passed as an argument to func.
redfuncReduction 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.
nChunksNumber 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 305 of file TProcessExecutor.hxx.

◆ MapReduce() [3/3]

template<class F , class R , class Cond >
auto ROOT::TProcessExecutor::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
funcFunction to be executed.
nTimesNumber of times function should be called.
Returns
A vector with the results of the function calls.
Parameters
redfuncReduction 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 265 of file TProcessExecutor.hxx.

◆ operator=()

TProcessExecutor & ROOT::TProcessExecutor::operator= ( const TProcessExecutor )
delete

◆ ReplyToFuncResult()

void ROOT::TProcessExecutor::ReplyToFuncResult ( TSocket s)
private

Reply to a worker who just sent a result.

If another argument to process exists, tell the worker. Otherwise send a shutdown order.

Definition at line 108 of file TProcessExecutor.cxx.

◆ ReplyToIdle()

void ROOT::TProcessExecutor::ReplyToIdle ( TSocket s)
private

Reply to a worker who is idle.

If another argument to process exists, tell the worker. Otherwise ask for a result

Definition at line 126 of file TProcessExecutor.cxx.

◆ Reset()

void ROOT::TProcessExecutor::Reset ( )
private

Reset TProcessExecutor's state.

Definition at line 97 of file TProcessExecutor.cxx.

◆ SetNWorkers()

void ROOT::TProcessExecutor::SetNWorkers ( unsigned  n)
inline

Definition at line 69 of file TProcessExecutor.hxx.

Member Data Documentation

◆ fNProcessed

unsigned ROOT::TProcessExecutor::fNProcessed
private

number of arguments already passed to the workers

Definition at line 96 of file TProcessExecutor.hxx.

◆ fNToProcess

unsigned ROOT::TProcessExecutor::fNToProcess
private

total number of arguments to pass to the workers

Definition at line 97 of file TProcessExecutor.hxx.

◆ fTaskType

ETask ROOT::TProcessExecutor::fTaskType = ETask::kNoTask
private

the kind of task that is being executed, if any

Definition at line 110 of file TProcessExecutor.hxx.

◆ TExecutorCRTP

friend ROOT::TProcessExecutor::TExecutorCRTP
private

Definition at line 38 of file TProcessExecutor.hxx.

Libraries for ROOT::TProcessExecutor:

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