Gwill / thread-pool

BS::thread_pool: a fast, lightweight, and easy-to-use C++17 thread pool library

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

DOI:10.5281/zenodo.4742687 arXiv:2105.00613 License: MIT Language: C++17 File size in bytes GitHub last commit GitHub repo stars Twitter @BarakShoshany Open in Visual Studio Code

BS::thread_pool: a fast, lightweight, and easy-to-use C++17 thread pool library

By Barak Shoshany
Email: baraksh@gmail.com
Website: https://baraksh.com/
GitHub: https://github.com/bshoshany

This is the complete documentation for v3.3.0 of the library, released on 2022-08-03.

Introduction

Motivation

Multithreading is essential for modern high-performance computing. Since C++11, the C++ standard library has included built-in low-level multithreading support using constructs such as std::thread. However, std::thread creates a new thread each time it is called, which can have a significant performance overhead. Furthermore, it is possible to create more threads than the hardware can handle simultaneously, potentially resulting in a substantial slowdown.

The library presented here contains a thread pool class, BS::thread_pool, which avoids these issues by creating a fixed pool of threads once and for all, and then continuously reusing the same threads to perform different tasks throughout the lifetime of the program. By default, the number of threads in the pool is equal to the maximum number of threads that the hardware can run in parallel.

The user submits tasks to be executed into a queue. Whenever a thread becomes available, it retrieves the next task from the queue and executes it. The pool automatically produces an std::future for each task, which allows the user to wait for the task to finish executing and/or obtain its eventual return value, if applicable. Threads and tasks are autonomously managed by the pool in the background, without requiring any input from the user aside from submitting the desired tasks.

The design of this package was guided by four important principles. First, compactness: the entire library consists of just one small self-contained header file, with no other components or dependencies. Second, portability: the package only utilizes the C++17 standard library, without relying on any compiler extensions or 3rd-party libraries, and is therefore compatible with any modern standards-conforming C++17 compiler on any platform. Third, ease of use: the package is extensively documented, and programmers of any level should be able to use it right out of the box.

The fourth and final guiding principle is performance: each and every line of code in this library was carefully designed with maximum performance in mind, and performance was tested and verified on a variety of compilers and platforms. Indeed, the library was originally designed for use in the author's own computationally-intensive scientific computing projects, running both on high-end desktop/laptop computers and high-performance computing nodes.

Other, more advanced multithreading libraries may offer more features and/or higher performance. However, they typically consist of a vast codebase with multiple components and dependencies, and involve complex APIs that require a substantial time investment to learn. This library is not intended to replace these more advanced libraries; instead, it was designed for users who don't require very advanced features, and prefer a simple and lightweight package that is easy to learn and use and can be readily incorporated into existing or new projects.

Overview of features

  • Fast:
    • Built from scratch with maximum performance in mind.
    • Suitable for use in high-performance computing nodes with a very large number of CPU cores.
    • Compact code, to reduce both compilation time and binary size.
    • Reusing threads avoids the overhead of creating and destroying them for individual tasks.
    • A task queue ensures that there are never more threads running in parallel than allowed by the hardware.
  • Lightweight:
    • Single header file: simply #include "BS_thread_pool.hpp" and you're all set!
    • Header-only: no need to install or build the library.
    • Self-contained: no external requirements or dependencies.
    • Portable: uses only the C++ standard library, and works with any C++17-compliant compiler.
    • Only ~340 lines of code, excluding comments and blank lines.
    • A stand-alone "light version" of the thread pool is also available in the BS_thread_pool_light.hpp header file, with only ~170 lines of code.
  • Easy to use:
    • Very simple operation, using a handful of member functions.
    • Every task submitted to the queue using the submit() member function automatically generates an std::future, which can be used to wait for the task to finish executing and/or obtain its eventual return value.
    • Optionally, tasks may also be submitted using the push_task() member function without generating a future, sacrificing convenience for even greater performance.
    • The code is thoroughly documented using Doxygen comments - not only the interface, but also the implementation, in case the user would like to make modifications.
    • The included test program BS_thread_pool_test.cpp can be used to perform exhaustive automated tests and benchmarks, and also serves as a comprehensive example of how to properly use the package.
  • Helper classes:
    • Automatically parallelize a loop into any number of parallel tasks using the parallelize_loop() member function, and track its execution using the BS::multi_future helper class.
    • Synchronize output to a stream from multiple threads in parallel using the BS::synced_stream helper class.
    • Easily measure execution time for benchmarking purposes using the BS::timer helper class.
  • Additional features:
    • Easily wait for all tasks in the queue to complete using the wait_for_tasks() member function.
    • Change the number of threads in the pool safely and on-the-fly as needed using the reset() member function.
    • Monitor the number of queued and/or running tasks using the get_tasks_queued(), get_tasks_running(), and get_tasks_total() member functions.
    • Freely pause and resume the pool using the pause(), unpause(), and is_paused() member functions. When paused, threads do not retrieve new tasks out of the queue.
    • Catch exceptions thrown by the submitted tasks.
    • Submit class member functions to the pool, either applied to a specific object or from within the object itself.
    • Under continuous and active development. Bug reports and feature requests are welcome, and should be made via GitHub issues.

Compiling and compatibility

This library should successfully compile on any C++17 standard-compliant compiler, on all operating systems and architectures for which such a compiler is available. Compatibility was verified with a 12-core / 24-thread AMD Ryzen 9 3900X CPU using the following compilers and platforms:

In addition, this library was tested on a Compute Canada node equipped with two 20-core / 40-thread Intel Xeon Gold 6148 CPUs (for a total of 40 cores and 80 threads), running CentOS Linux 7.9.2009, using GCC v12.1.1.

The test program BS_thread_pool_test.cpp was compiled without warnings (with the warning flags -Wall -Wextra -Wconversion -Wsign-conversion -Wpedantic -Weffc++ -Wshadow in GCC/Clang and /W4 in MSVC), executed, and successfully completed all automated tests and benchmarks using all of the compilers and systems mentioned above.

As this library requires C++17 features, the code must be compiled with C++17 support:

  • For Clang or GCC, use the -std=c++17 flag. On Linux, you will also need to use the -pthread flag to enable the POSIX threads library.
  • For MSVC, use /std:c++17, and preferably also /permissive- to ensure standards conformance.

For maximum performance, it is recommended to compile with all available compiler optimizations:

  • For Clang or GCC, use the -O3 flag.
  • For MSVC, use /O2.

As an example, to compile the test program BS_thread_pool_test.cpp with warnings and optimizations, it is recommended to use the following commands:

  • On Linux with GCC: g++ BS_thread_pool_test.cpp -std=c++17 -O3 -Wall -Wextra -Wconversion -Wsign-conversion -Wpedantic -Weffc++ -Wshadow -pthread -o BS_thread_pool_test
  • On Linux with Clang: replace g++ with clang++.
  • On Windows with GCC or Clang: replace -o BS_thread_pool_test with -o BS_thread_pool_test.exe and remove -pthread.
  • On Windows with MSVC: cl BS_thread_pool_test.cpp /std:c++17 /permissive- /O2 /W4 /EHsc /Fe:BS_thread_pool_test.exe

Installing using vcpkg

If you are using the vcpkg C/C++ library manager, you can easily download and install this package with the following commands.

On Linux/macOS:

./vcpkg install bshoshany-thread-pool

On Windows:

.\vcpkg install bshoshany-thread-pool:x86-windows bshoshany-thread-pool:x64-windows

The thread pool will then be available automatically in the build system you integrated vcpkg with (e.g. MSBuild or CMake). Simply write #include "BS_thread_pool.hpp" in any project to use the thread pool, without having to copy to file into the project first. I will update the vcpkg port with each new release, so it will be updated automatically when you run vcpkg upgrade.

Please see the vcpkg repository for more information on how to use vcpkg.

Installing using Conan

If you are using the Conan C/C++ package manager, please refer to this package's page on ConanCenter to learn how to use Conan to include this package in your project with various build systems.

Getting started

Including the library

If you are not using a C++ library manager (such as vcpkg), simply download the latest release from the GitHub repository, place the single header file BS_thread_pool.hpp in the desired folder, and include it in your program:

#include "BS_thread_pool.hpp"

The thread pool will now be accessible via the BS::thread_pool class.

Constructors

The default constructor creates a thread pool with as many threads as the hardware can handle concurrently, as reported by the implementation via std::thread::hardware_concurrency(). This is usually determined by the number of cores in the CPU. If a core is hyperthreaded, it will count as two threads. For example:

// Constructs a thread pool with as many threads as available in the hardware.
BS::thread_pool pool;

Optionally, a number of threads different from the hardware concurrency can be specified as an argument to the constructor. However, note that adding more threads than the hardware can handle will not improve performance, and in fact will most likely hinder it. This option exists in order to allow using less threads than the hardware concurrency, in cases where you wish to leave some threads available for other processes. For example:

// Constructs a thread pool with only 12 threads.
BS::thread_pool pool(12);

If your program's main thread only submits tasks to the thread pool and waits for them to finish, and does not perform any computationally intensive tasks on its own, then it is recommended to use the default value for the number of threads. This ensures that all of the threads available in the hardware will be put to work while the main thread waits.

However, if your main thread does perform computationally intensive tasks on its own, then it is recommended to use the value std::thread::hardware_concurrency() - 1 for the number of threads. In this case, the main thread plus the thread pool will together take up exactly all the threads available in the hardware.

Getting and resetting the number of threads in the pool

The member function get_thread_count() returns the number of threads in the pool. This will be equal to std::thread::hardware_concurrency() if the default constructor was used.

It is generally unnecessary to change the number of threads in the pool after it has been created, since the whole point of a thread pool is that you only create the threads once. However, if needed, this can be done, safely and on-the-fly, using the reset() member function.

reset() will wait for all currently running tasks to be completed, but will leave the rest of the tasks in the queue. Then it will destroy the thread pool and create a new one with the desired new number of threads, as specified in the function's argument (or the hardware concurrency if no argument is given). The new thread pool will then resume executing the tasks that remained in the queue and any new submitted tasks.

Finding the version of the package

If desired, the version of this package may be read during compilation time from the macro BS_THREAD_POOL_VERSION. The value will be a string containing the version number and release date. For example:

std::cout << "Thread pool library version is " << BS_THREAD_POOL_VERSION << ".\n";

Sample output:

Thread pool library version is v3.1.0 (2022-07-13).

This can be used, for example, to allow the same code to work with several incompatible versions of the library.

Submitting and waiting for tasks

Submitting tasks to the queue with futures

A task can be any function, with zero or more arguments, and with or without a return value. Once a task has been submitted to the queue, it will be executed as soon as a thread becomes available. Tasks are executed in the order that they were submitted (first-in, first-out).

The member function submit() is used to submit tasks to the queue. The first argument is the function to execute, and the rest of the arguments are the arguments to pass to the function, if any. The return value is an std::future associated to the task. For example:

// Submit a task without arguments to the queue, and get a future for it.
auto my_future = pool.submit(task);
// Submit a task with one argument to the queue, and get a future for it.
auto my_future = pool.submit(task, arg);
// Submit a task with two arguments to the queue, and get a future for it.
auto my_future = pool.submit(task, arg1, arg2);

If the submitted function has a return value of type T, then the future will be of type std::future<T>, and will be set to the return value when the function finishes its execution. If the submitted function does not have a return value, then the future will be an std::future<void>, which will not return any value but may still be used to wait for the function to finish.

Using auto for the return value of submit() means the compiler will automatically detect which instance of the template std::future to use. However, specifying the particular type std::future<T>, as in the examples below, is recommended for increased readability.

To wait until the task finishes, use the member function wait() of the future. To obtain the return value, use the member function get(), which will also automatically wait for the task to finish if it hasn't yet. For example:

// Submit a task and get a future.
auto my_future = pool.submit(task);
// Do some other stuff while the task is executing.
do_stuff();
// Get the task's return value from the future, waiting for it to finish running if needed.
auto my_return_value = my_future.get();

Here are some more concrete examples. The following program will print out 42:

#include "BS_thread_pool.hpp"

int main()
{
    BS::thread_pool pool;
    std::future<int> my_future = pool.submit([] { return 42; });
    std::cout << my_future.get();
}

Here we used a lambda expression to quickly define the function on-the-fly. However, we can also use a previously-defined function:

#include "BS_thread_pool.hpp"

int the_answer()
{
    return 42;
}

int main()
{
    BS::thread_pool pool;
    std::future<int> my_future = pool.submit(the_answer);
    std::cout << my_future.get();
}

The following is an example of submitting a function with arguments:

#include "BS_thread_pool.hpp"

int multiply(const int a, const int b)
{
    return a * b;
}

int main()
{
    BS::thread_pool pool;
    std::future<int> my_future = pool.submit(multiply, 6, 7);
    std::cout << my_future.get();
}

Finally, here is an example of submitting a function with no return value and then using the future to wait for it to finish executing:

#include "BS_thread_pool.hpp"

void sleep()
{
    std::this_thread::sleep_for(std::chrono::milliseconds(1000));
}

int main()
{
    BS::thread_pool pool;
    std::future<void> my_future = pool.submit(sleep);
    std::cout << "Waiting... ";
    my_future.wait();
    std::cout << "Done.";
}

Here, the command std::this_thread::sleep_for(std::chrono::milliseconds(1000)) instructs the thread to sleep for 1 second.

Submitting tasks to the queue without futures

Usually, it is best to submit a task to the queue using submit(). This allows you to wait for the task to finish and/or get its return value later. However, sometimes a future is not needed, for example when you just want to "set and forget" a certain task, or if the task already communicates with the main thread or with other tasks without using futures, such as via condition variables. In such cases, you may wish to avoid the overhead involved in assigning a future to the task in order to increase performance.

The member function push_task() allows you to submit a task to the queue without generating a future for it. The task can have any number of arguments, but it cannot have a return value. For example:

// Submit a task without arguments or return value to the queue.
pool.push_task(task);
// Submit a task with one argument and no return value to the queue.
pool.push_task(task, arg);
// Submit a task with two arguments and no return value to the queue.
pool.push_task(task, arg1, arg2);

Warning! Since push_task() does not return a future, there is no built-in way for the user to know when the task finishes executing. You must use either wait_for_tasks() (see below), or some other method such as condition variables, to ensure that the task finishes executing before trying to use anything that depends on its output. Otherwise, bad things will happen!

Manually waiting for all tasks to complete

To wait for a single submitted task to complete, use submit() and then use the wait() or get() member functions of the obtained future. However, in cases where you need to wait until all submitted tasks finish their execution, or if the tasks have been submitted without futures using push_task(), you can use the member function wait_for_tasks().

Consider, for example, the following code:

#include "BS_thread_pool.hpp"

int main()
{
    BS::thread_pool pool(5);
    int squares[100];
    for (int i = 0; i < 100; ++i)
        pool.push_task(
            [&squares, i]
            {
                std::this_thread::sleep_for(std::chrono::milliseconds(50));
                squares[i] = i * i;
            });
    std::cout << squares[50];
}

The output will most likely be garbage, since the task that modifies squares[50] has not yet finished executing by the time we try to access that element - it's still waiting in the queue. One solution would be to use submit() instead of push_task(), but perhaps we don't want the overhead of generating 100 different futures. Instead, simply adding the line

pool.wait_for_tasks();

after the for loop will ensure - as efficiently as possible - that all tasks have finished running before we attempt to access any elements of the array squares, and the code will print out the value 2500 as expected.

Note, however, that wait_for_tasks() will wait for all the tasks in the queue, including those that are unrelated to the for loop. Using parallelize_loop() would make much more sense in this particular case, as it will allow waiting only for the tasks related to the loop.

Submitting class member functions to the queue

Consider the following program:

#include "BS_thread_pool.hpp"

BS::thread_pool pool;

class flag_class
{
public:
    bool get_flag() const
    {
        return flag;
    }

    void set_flag(const bool arg)
    {
        flag = arg;
    }

private:
    bool flag = false;
};

int main()
{
    flag_class flag_object;
    flag_object.set_flag(true);
    std::cout << std::boolalpha << flag_object.get_flag() << '\n';
}

This program creates a new object flag_object of the class flag_class, sets the flag to true using the member function set_flag(), and then prints out the flag's value. But what if you want to submit the member function set_flag() as a task to the thread pool?

To submit member functions to the pool, use the following general syntax:

pool.push_task(&class::function, &object, args);

The same syntax also works with submit(). Note that, in the second argument, you must specify the object on which the member function will be executed (unless it's a static member function, in which case you just submit it like any other function). Also note that both the first and second arguments must be pointers, so they must be preceded by the & operator.

If you remove the & from the first argument, the code won't work with most compilers, and if you remove the & from the second argument, the function will act on a copy of the object, rather than on the object itself, so any changes made to the object will not be saved. Therefore, it's important to ensure that both arguments are pointers.

To make the above program submit the member function set_flag() to the thread pool, simply replace the line:

flag_object.set_flag(true);

with:

pool.push_task(&flag_class::set_flag, &flag_object, true);
pool.wait_for_tasks();

Here the class is flag_class, the name of the function is set_flag, the object we want the function to act on is flag_object, and the argument to pass to the function is true.

Another thing you might want to do is call a member function from within the object itself, that is, from another member function. This follows a similar syntax, except that you don't need to specify the class, and you use this to get a pointer to the current object (no & necessary, since this is already a pointer). Here is an example, this time using submit():

#include "BS_thread_pool.hpp"

BS::thread_pool pool;

class flag_class
{
public:
    bool get_flag() const
    {
        return flag;
    }

    void set_flag(const bool arg)
    {
        flag = arg;
    }

    void set_flag_to_true()
    {
        pool.submit(&flag_class::set_flag, this, true).wait();
    }

private:
    bool flag = false;
};

int main()
{
    flag_class flag_object;
    flag_object.set_flag_to_true();
    std::cout << std::boolalpha << flag_object.get_flag() << '\n';
}

Parallelizing loops

One of the most common and effective methods of parallelization is splitting a loop into smaller loops and running them in parallel. It is most effective in "embarrassingly parallel" computations, such as vector or matrix operations, where each iteration of the loop is completely independent of every other iteration. For example, if we are summing up two vectors of 1000 elements each, and we have 10 threads, we could split the summation into 10 blocks of 100 elements each, and run all the blocks in parallel, potentially increasing performance by up to a factor of 10.

BS::thread_pool can automatically parallelize loops. To see how this works, consider the following generic loop:

for (T i = start; i < end; ++i)
    do_something(i);

where:

  • T is any signed or unsigned integer type.
  • The loop is over the range [start, end), i.e. inclusive of start but exclusive of end.
  • do_something() is an operation performed for each loop index i, such as modifying an array with end - start elements.

This loop may be automatically parallelized and submitted to the thread pool's queue using the member function parallelize_loop(), which has the follows syntax:

pool.parallelize_loop(start, end, loop, num_blocks);

where:

  • start is the first index in the range.
    • This argument can be omitted, in which case it is assumed that the loop starts at 0. That is, parallelize_loop(end, loop, num_blocks) is equivalent to parallelize_loop(0, end, loop, num_blocks).
  • end is the index after the last index in the range, such that the full range is [start, end). In other words, the loop will be equivalent to the one above if start and end are the same.
    • start and end should both be integers, but they need not be of the same integer type. parallelize_loop() will automatically determine the best type to use for the loop indices.
  • loop() is any function that takes two indices, a, and b, and executes only the portion of the loop in the range [a, b). Typically, loop() will include a for loop of the form for (T i = a; i < b; ++i).
  • num_blocks is the number of blocks of the form [a, b) to split the loop into. For example, if the range is [0, 9) and there are 3 blocks, then the blocks will be the ranges [0, 3), [3, 6), and [6, 9). If possible, the blocks will be equal in size; otherwise, the last block may be a bit longer.
    • This argument can be omitted, in which case the number of blocks will be the number of threads in the pool.

Each block will be submitted to the thread pool's queue as a separate task. Therefore, a loop that is split into 3 blocks will be split into 3 individual tasks, which may run in parallel. If there is only one block, then the entire loop will run as one task, and no parallelization will take place.

To parallelize the generic loop above, we use the following code:

auto loop = [](const T a, const T b)
{
    for (T i = a; i < b; ++i)
        do_something(i);
};
BS::multi_future<void> loop_future = pool.parallelize_loop(start, end, loop, num_blocks);
loop_future.wait();

Here we defined loop() as a lambda function. Of course, loop() could also be defined as a lambda within the call to parallelize_loop() itself, as in the examples below; or it could be any ordinary function, but a lambda is preferred since one typically would like to capture some of the surrounding variables.

parallelize_loop() returns an object of the helper class template BS::multi_future. Each of the num_blocks blocks will have an std::future assigned to it, and all these futures will be stored inside the returned BS::multi_future object. When loop_future.wait() is called, the main thread will wait until all tasks generated by parallelize_loop() finish executing, and only those tasks - not any other tasks that also happen to be in the queue. This is essentially the role of the BS::multi_future class: to wait for a specific group of tasks, in this case the tasks running the loop blocks.

What value should you use for num_blocks? Omitting this argument, so that the number of blocks will be equal to the number of threads in the pool, is typically a good choice. For best performance, it is recommended to do your own benchmarks to find the optimal number of blocks for each loop (you can use the BS::timer helper class - see below). Using less tasks than there are threads may be preferred if you are also running other tasks in parallel. Using more tasks than there are threads may improve performance in some cases.

As a simple example, the following code calculates and prints the squares of all integers from 0 to 99:

#include "BS_thread_pool.hpp"

int main()
{
    int squares[100];
    for (int i = 0; i < 100; ++i)
    {
        squares[i] = i * i;
        std::cout << i << "^2 = " << squares[i] << "   ";
    }
}

We can parallelize it as follows:

#include "BS_thread_pool.hpp"

int main()
{
    BS::thread_pool pool(10);
    int squares[100];
    pool.parallelize_loop(100,
                          [&squares](const int a, const int b)
                          {
                              for (int i = a; i < b; ++i)
                                  squares[i] = i * i;
                          })
        .wait();
    for (int i = 0; i < 100; ++i)
        std::cout << i << "^2 = " << squares[i] << "   ";
}

Since there are 10 threads, and we omitted the num_blocks argument, the loop will be divided into 10 blocks, each calculating 10 squares. Also, since the loop starts from 0, we did not need to specify the first index.

In this example, instead of storing the BS::multi_future object and then using it to wait, we simply called the wait() member function directly on the temporary object returned by parallelize_loop() without storing it anywhere. This is a convenient shortcut when we have nothing else to do while waiting.

Notice that here we parallelized the calculation of the squares, but we did not parallelize printing the results. This is for two reasons:

  1. We want to print out the squares in ascending order, and we have no guarantee that the blocks will be executed in the correct order. This is very important; you must never expect that the parallelized loop will execute at the same order as the non-parallelized loop.
  2. If we did print out the squares from within the parallel tasks, we would get a huge mess, since all 10 blocks would print to the standard output at once. Later we will see how to synchronize printing to a stream from multiple tasks at the same time.

Loops with return values

Usually, parallelize_loop() should take functions with no return values. This is because the function will be executed once for each block, but the blocks are managed by the thread pool, so there's limited usability in returning one value per block. However, for the case where this is desired, such as for summation or some sorting algorithms, parallelize_loop() does accept functions with return values, in which case it returns a BS::multi_future<T> object where T is the type of the return values.

Here's an example of summing all the numbers from 1 to 100:

#include "BS_thread_pool.hpp"

int main()
{
    BS::thread_pool pool;
    BS::multi_future<int> mf = pool.parallelize_loop(1, 101,
                                                     [](const int a, const int b)
                                                     {
                                                         int block_total = 0;
                                                         for (int i = a; i < b; ++i)
                                                             block_total += i;
                                                         return block_total;
                                                     });
    std::vector<int> totals = mf.get();
    int sum = 0;
    for (const int t : totals)
        sum += t;
    std::cout << sum;
}

Calling get() on a BS::multi_future<T> object returns an std::vector<T> with the values obtained from each future. In this case, the values will be the partial sums from each block, so when we add them up, we will get the total sum.

Parallelizing loops without futures

Just as in the case of push_task() vs. submit(), sometimes you may want to parallelize a loop, but you don't need it to return a BS::multi_future. In this case, you can save the overhead of generating the futures (which can be significant, depending on the number of blocks) by using push_loop() instead of parallelize_loop(), with the same arguments.

For example, you could also run the loop of squares example above as follows:

#include "BS_thread_pool.hpp"

int main()
{
    BS::thread_pool pool(10);
    int squares[100];
    pool.push_loop(100,
                   [&squares](const int a, const int b)
                   {
                       for (int i = a; i < b; ++i)
                           squares[i] = i * i;
                   });
    pool.wait_for_tasks();
    for (int i = 0; i < 100; ++i)
        std::cout << i << "^2 = " << squares[i] << "   ";
}

As with parallelize_loop(), the first argument can be omitted if the start index is 0, and the last argument can be omitted if the number of blocks should be equal to the number of threads.

Warning! Since push_loop() does not return a BS::multi_future, there is no built-in way for the user to know when the loop finishes executing. You must use either wait_for_tasks(), or some other method such as condition variables, to ensure that the loop finishes executing before trying to use anything that depends on its output. Otherwise, bad things will happen!

Helper classes

Synchronizing printing to an output stream

When printing to an output stream from multiple threads in parallel, the output may become garbled. For example, consider this code:

#include "BS_thread_pool.hpp"

int main()
{
    BS::thread_pool pool;
    for (size_t i = 1; i <= 5; ++i)
        pool.push_task([i] { std::cout << "Task no. " << i << " executing.\n"; });
}

The output may look as follows:

Task no. Task no. 2Task no. 5 executing.
Task no.  executing.
Task no. 1 executing.
4 executing.
3 executing.

The reason is that, although each individual insertion to std::cout is thread-safe, there is no mechanism in place to ensure subsequent insertions from the same thread are printed contiguously.

The helper class BS::synced_stream is designed to eliminate such synchronization issues. The constructor takes one optional argument, specifying the output stream to print to. If no argument is supplied, std::cout will be used:

// Construct a synced stream that will print to std::cout.
BS::synced_stream sync_out;
// Construct a synced stream that will print to the output stream my_stream.
BS::synced_stream sync_out(my_stream);

The member function print() takes an arbitrary number of arguments, which are inserted into the stream one by one, in the order they were given. println() does the same, but also prints a newline character \n at the end, for convenience. A mutex is used to synchronize this process, so that any other calls to print() or println() using the same BS::synced_stream object must wait until the previous call has finished.

As an example, this code:

#include "BS_thread_pool.hpp"

int main()
{
    BS::synced_stream sync_out;
    BS::thread_pool pool;
    for (size_t i = 1; i <= 5; ++i)
        pool.push_task([i, &sync_out] { sync_out.println("Task no. ", i, " executing."); });
}

Will print out:

Task no. 1 executing.
Task no. 2 executing.
Task no. 3 executing.
Task no. 4 executing.
Task no. 5 executing.

Warning: Always create the BS::synced_stream object before the BS::thread_pool object, as we did in this example. When the BS::thread_pool object goes out of scope, it waits for the remaining tasks to be executed. If the BS::synced_stream object goes out of scope before the BS::thread_pool object, then any tasks using the BS::synced_stream will crash. Since objects are destructed in the opposite order of construction, creating the BS::synced_stream object before the BS::thread_pool object ensures that the BS::synced_stream is always available to the tasks, even while the pool is destructing.

Most stream manipulators defined in the headers <ios> and <iomanip>, such as std::setw (set the character width of the next output), std::setprecision (set the precision of floating point numbers), and std::fixed (display floating point numbers with a fixed number of digits), can be passed to print() and println() just as you would pass them to a stream.

The only exceptions are the flushing manipulators std::endl and std::flush, which will not work because the compiler will not be able to figure out which template specializations to use. Instead, use BS::synced_stream::endl and BS::synced_stream::flush. Here is an example:

#include "BS_thread_pool.hpp"
#include <cmath>
#include <iomanip>

int main()
{
    BS::synced_stream sync_out;
    BS::thread_pool pool;
    sync_out.print(std::setprecision(10), std::fixed);
    for (size_t i = 1; i <= 10; ++i)
        pool.push_task([i, &sync_out] { sync_out.print("The square root of ", std::setw(2), i, " is ", std::sqrt(i), ".", BS::synced_stream::endl); });
}

Handling multiple futures at once

The helper class template BS::multi_future<T>, already introduced in the context of parallelize_loop(), provides a convenient way to collect and access groups of futures. This class works similarly to STL containers such as std::vector:

  • When you create a new object, either use the default constructor to create an empty object and add futures to it later, or pass the desired number of futures to the constructor in advance.
  • Use the [] operator to access the future at a specific index, or the push_back() member function to append a new future to the list.
  • The size() member function tells you how many futures are currently stored in the object.
  • Once all the futures are stored, you can use wait() to wait for all of them at once or get() to get an std::vector<T> with the results from all of them.

Aside from using BS::multi_future to track the execution of parallelized loops, it can also be used whenever you have several different groups of tasks and you want to track the execution of each group individually. Here's a simple example:

#include "BS_thread_pool.hpp"
#include <cmath>

BS::synced_stream sync_out;
BS::thread_pool pool;

double power(const double i, const double j)
{
    std::this_thread::sleep_for(std::chrono::milliseconds(10 * pool.get_thread_count()));
    return std::pow(i, j);
}

void print_vector(const std::vector<double>& vec)
{
    for (const double i : vec)
        sync_out.print(i, ' ');
    sync_out.println();
}

int main()
{
    constexpr size_t n = 100;

    // First group of tasks: calculate n squares.
    // Here we create an empty BS::multi_future object, and append futures to it via push_back().
    BS::multi_future<double> mf_squares;
    for (int i = 0; i < n; ++i)
        mf_squares.push_back(pool.submit(power, i, 2));

    // Second group of tasks: calculate n cubes.
    // In this case, we create a BS::multi_future object of the desired size in advance, and store the futures via the [] operator. This is faster since there will be no memory reallocations, but also more prone to errors.
    BS::multi_future<double> mf_cubes(n);
    for (int i = 0; i < n; ++i)
        mf_cubes[i] = pool.submit(power, i, 3);

    // Both groups are now queued, but it will take some time until they all execute.

    /// ...
    /// Do some stuff while the first group of tasks executes...
    /// ...

    // Get and print the results from the first group.
    sync_out.println("Squares:");
    print_vector(mf_squares.get());

    /// ...
    /// Do other stuff while the second group of tasks executes...
    /// ...

    // Get and print the results from the second group.
    sync_out.println("Cubes:");
    print_vector(mf_cubes.get());
}

In this example, we simulate complicated tasks by having each task wait for a bit before returning its result. We collect the futures of the tasks submitted within each group into two separate BS::multi_future<double> objects. mf_squares holds the results from the first group, and mf_cubes holds the results from the second group. Now we can wait for and/or get the results from mf_squares whenever is convenient, and separately wait for and/or get the results from mf_cubes at another time.

Measuring execution time

If you are using a thread pool, then your code is most likely performance-critical. Achieving maximum performance requires performing a considerable amount of benchmarking to determine the optimal settings and algorithms. Therefore, it is important to be able to measure the execution time of various computations and operations under different conditions.

The helper class BS::timer provides a simple way to measure execution time. It is very straightforward to use:

  1. Create a new BS::timer object.
  2. Immediately before you execute the computation that you want to time, call the start() member function.
  3. Immediately after the computation ends, call the stop() member function.
  4. Use the member function ms() to obtain the elapsed time for the computation in milliseconds.

For example:

BS::timer tmr;
tmr.start();
do_something();
tmr.stop();
std::cout << "The elapsed time was " << tmr.ms() << " ms.\n";

A practical application of the BS::timer class can be found in the benchmark portion of the test program BS_thread_pool_test.cpp.

Other features

Monitoring the tasks

Sometimes you may wish to monitor what is happening with the tasks you submitted to the pool. This may be done using three member functions:

  • get_tasks_queued() gets the number of tasks currently waiting in the queue to be executed by the threads.
  • get_tasks_running() gets the number of tasks currently being executed by the threads.
  • get_tasks_total() gets the total number of unfinished tasks: either still in the queue, or running in a thread.
  • Note that get_tasks_total() == get_tasks_queued() + get_tasks_running().

These functions are demonstrated in the following program:

#include "BS_thread_pool.hpp"

BS::synced_stream sync_out;
BS::thread_pool pool(4);

void sleep_half_second(const size_t i)
{
    std::this_thread::sleep_for(std::chrono::milliseconds(500));
    sync_out.println("Task ", i, " done.");
}

void monitor_tasks()
{
    sync_out.println(pool.get_tasks_total(), " tasks total, ", pool.get_tasks_running(), " tasks running, ", pool.get_tasks_queued(), " tasks queued.");
}

int main()
{
    for (size_t i = 0; i < 12; ++i)
        pool.push_task(sleep_half_second, i);
    monitor_tasks();
    std::this_thread::sleep_for(std::chrono::milliseconds(750));
    monitor_tasks();
    std::this_thread::sleep_for(std::chrono::milliseconds(500));
    monitor_tasks();
    std::this_thread::sleep_for(std::chrono::milliseconds(500));
    monitor_tasks();
}

Assuming you have at least 4 hardware threads (so that 4 tasks can run concurrently), the output should be similar to:

12 tasks total, 0 tasks running, 12 tasks queued.
Task 0 done.
Task 1 done.
Task 2 done.
Task 3 done.
8 tasks total, 4 tasks running, 4 tasks queued.
Task 4 done.
Task 5 done.
Task 6 done.
Task 7 done.
4 tasks total, 4 tasks running, 0 tasks queued.
Task 8 done.
Task 9 done.
Task 10 done.
Task 11 done.
0 tasks total, 0 tasks running, 0 tasks queued.

Pausing the workers

Sometimes you may wish to temporarily pause the execution of tasks, or perhaps you want to submit tasks to the queue in advance and only start executing them at a later time. You can do this using the member functions pause(), unpause(), and is_paused().

When you call pause(), the workers will temporarily stop retrieving new tasks out of the queue. However, any tasks already executed will keep running until they are done, since the thread pool has no control over the internal code of your tasks. If you need to pause a task in the middle of its execution, you must do that manually by programming your own pause mechanism into the task itself. To resume retrieving tasks, call unpause(). To check whether the pool is currently paused, call is_paused().

Here is an example:

#include "BS_thread_pool.hpp"

BS::synced_stream sync_out;
BS::thread_pool pool(4);

void sleep_half_second(const size_t i)
{
    std::this_thread::sleep_for(std::chrono::milliseconds(500));
    sync_out.println("Task ", i, " done.");
}

void check_if_paused()
{
    if (pool.is_paused())
        sync_out.println("Pool paused.");
    else
        sync_out.println("Pool unpaused.");
}

int main()
{
    for (size_t i = 0; i < 8; ++i)
        pool.push_task(sleep_half_second, i);
    sync_out.println("Submitted 8 tasks.");
    std::this_thread::sleep_for(std::chrono::milliseconds(250));
    pool.pause();
    check_if_paused();
    std::this_thread::sleep_for(std::chrono::milliseconds(1000));
    sync_out.println("Still paused...");
    std::this_thread::sleep_for(std::chrono::milliseconds(1000));
    for (size_t i = 8; i < 12; ++i)
        pool.push_task(sleep_half_second, i);
    sync_out.println("Submitted 4 more tasks.");
    sync_out.println("Still paused...");
    std::this_thread::sleep_for(std::chrono::milliseconds(1000));
    pool.unpause();
    check_if_paused();
}

Assuming you have at least 4 hardware threads, the output should be similar to:

Submitted 8 tasks.
Pool paused.
Task 0 done.
Task 1 done.
Task 2 done.
Task 3 done.
Still paused...
Submitted 4 more tasks.
Still paused...
Pool unpaused.
Task 4 done.
Task 5 done.
Task 6 done.
Task 7 done.
Task 8 done.
Task 9 done.
Task 10 done.
Task 11 done.

Here is what happened. We initially submitted a total of 8 tasks to the queue. Since we waited for 250ms before pausing, the first 4 tasks have already started running, so they kept running until they finished. While the pool was paused, we submitted 4 more tasks to the queue, but they just waited at the end of the queue. When we unpaused, the remaining 4 initial tasks were executed, followed by the 4 new tasks.

While the workers are paused, wait_for_tasks() will wait for the running tasks instead of all tasks (otherwise it would wait forever). This is demonstrated by the following program:

#include "BS_thread_pool.hpp"

BS::synced_stream sync_out;
BS::thread_pool pool(4);

void sleep_half_second(const size_t i)
{
    std::this_thread::sleep_for(std::chrono::milliseconds(500));
    sync_out.println("Task ", i, " done.");
}

void check_if_paused()
{
    if (pool.is_paused())
        sync_out.println("Pool paused.");
    else
        sync_out.println("Pool unpaused.");
}

int main()
{
    for (size_t i = 0; i < 8; ++i)
        pool.push_task(sleep_half_second, i);
    sync_out.println("Submitted 8 tasks. Waiting for them to complete.");
    pool.wait_for_tasks();
    for (size_t i = 8; i < 20; ++i)
        pool.push_task(sleep_half_second, i);
    sync_out.println("Submitted 12 more tasks.");
    std::this_thread::sleep_for(std::chrono::milliseconds(250));
    pool.pause();
    check_if_paused();
    sync_out.println("Waiting for the ", pool.get_tasks_running(), " running tasks to complete.");
    pool.wait_for_tasks();
    sync_out.println("All running tasks completed. ", pool.get_tasks_queued(), " tasks still queued.");
    std::this_thread::sleep_for(std::chrono::milliseconds(1000));
    sync_out.println("Still paused...");
    std::this_thread::sleep_for(std::chrono::milliseconds(1000));
    sync_out.println("Still paused...");
    std::this_thread::sleep_for(std::chrono::milliseconds(1000));
    pool.unpause();
    check_if_paused();
    std::this_thread::sleep_for(std::chrono::milliseconds(250));
    sync_out.println("Waiting for the remaining ", pool.get_tasks_total(), " tasks (", pool.get_tasks_running(), " running and ", pool.get_tasks_queued(), " queued) to complete.");
    pool.wait_for_tasks();
    sync_out.println("All tasks completed.");
}

The output should be similar to:

Submitted 8 tasks. Waiting for them to complete.
Task 0 done.
Task 1 done.
Task 2 done.
Task 3 done.
Task 4 done.
Task 5 done.
Task 6 done.
Task 7 done.
Submitted 12 more tasks.
Pool paused.
Waiting for the 4 running tasks to complete.
Task 8 done.
Task 9 done.
Task 10 done.
Task 11 done.
All running tasks completed. 8 tasks still queued.
Still paused...
Still paused...
Pool unpaused.
Waiting for the remaining 8 tasks (4 running and 4 queued) to complete.
Task 12 done.
Task 13 done.
Task 14 done.
Task 15 done.
Task 16 done.
Task 17 done.
Task 18 done.
Task 19 done.
All tasks completed.

The first wait_for_tasks(), which was called while the pool was not paused, waited for all 8 tasks, both running and queued. The second wait_for_tasks(), which was called after pausing the pool, only waited for the 4 running tasks, while the other 8 tasks remained queued, and were not executed since the pool was paused. Finally, the third wait_for_tasks(), which was called after unpausing the pool, waited for the remaining 8 tasks, both running and queued.

Warning: If the thread pool is destroyed while paused, any tasks still in the queue will never be executed!

Exception handling

submit() catches any exceptions thrown by the submitted task and forwards them to the corresponding future. They can then be caught when invoking the get() member function of the future. For example:

#include "BS_thread_pool.hpp"

BS::synced_stream sync_out;
BS::thread_pool pool;

double inverse(const double x)
{
    if (x == 0)
        throw std::runtime_error("Division by zero!");
    else
        return 1 / x;
}

int main()
{
    constexpr double num = 0;
    std::future<double> my_future = pool.submit(inverse, num);
    try
    {
        const double result = my_future.get();
        sync_out.println("The inverse of ", num, " is ", result, ".");
    }
    catch (const std::exception& e)
    {
        sync_out.println("Caught exception: ", e.what());
    }
}

The output will be:

Caught exception: Division by zero!

However, if you change num to any non-zero number, no exceptions will be thrown and the inverse will be printed.

It is important to note that wait() does not throw any exceptions; only get() does. Therefore, even if your task does not return anything, i.e. your future is an std::future<void>, you must still use get() on the future obtained from it if you want to catch exceptions thrown by it. Here is an example:

#include "BS_thread_pool.hpp"

BS::synced_stream sync_out;
BS::thread_pool pool;

void print_inverse(const double x)
{
    if (x == 0)
        throw std::runtime_error("Division by zero!");
    else
        sync_out.println("The inverse of ", x, " is ", 1 / x, ".");
}

int main()
{
    constexpr double num = 0;
    std::future<void> my_future = pool.submit(print_inverse, num);
    try
    {
        my_future.get();
    }
    catch (const std::exception& e)
    {
        sync_out.println("Caught exception: ", e.what());
    }
}

When using BS::multi_future to handle multiple futures at once, exception handling works the same way: if any of the futures may throw exceptions, you may catch these exceptions when calling get(), even in the case of BS::multi_future<void>.

Testing the package

The included file BS_thread_pool_test.cpp will perform automated tests of all aspects of the package, and perform simple benchmarks. The output will be printed both to std::cout and to a file named BS_thread_pool_test-yyyy-mm-dd_hh.mm.ss.log based on the current date and time. In addition, the code is thoroughly documented, and is meant to serve as an extensive example of how to properly use the package.

Please make sure to:

  1. Compile BS_thread_pool_test.cpp with optimization flags enabled (e.g. -O3 on GCC / Clang or /O2 on MSVC).
  2. Run the test without any other applications, especially multithreaded applications, running in parallel.

If any of the tests fail, please submit a bug report including the exact specifications of your system (OS, CPU, compiler, etc.) and the generated log file.

Automated tests

A sample output of a successful run of the automated tests is as follows:

BS::thread_pool: a fast, lightweight, and easy-to-use C++17 thread pool library
(c) 2022 Barak Shoshany (baraksh@gmail.com) (http://baraksh.com)
GitHub: https://github.com/bshoshany/thread-pool

Thread pool library version is v3.3.0 (2022-08-03).
Hardware concurrency is 24.
Generating log file: BS_thread_pool_test-2022-08-03_12.32.04.log.

Important: Please do not run any other applications, especially multithreaded applications, in parallel with this test!

====================================
Checking that the constructor works:
====================================
Checking that the thread pool reports a number of threads equal to the hardware concurrency...
Expected: 24, obtained: 24 -> PASSED!
Checking that the manually counted number of unique thread IDs is equal to the reported number of threads...
Expected: 24, obtained: 24 -> PASSED!

============================
Checking that reset() works:
============================
Checking that after reset() the thread pool reports a number of threads equal to half the hardware concurrency...
Expected: 12, obtained: 12 -> PASSED!
Checking that after reset() the manually counted number of unique thread IDs is equal to the reported number of threads...
Expected: 12, obtained: 12 -> PASSED!
Checking that after a second reset() the thread pool reports a number of threads equal to the hardware concurrency...
Expected: 24, obtained: 24 -> PASSED!
Checking that after a second reset() the manually counted number of unique thread IDs is equal to the reported number of threads...
Expected: 24, obtained: 24 -> PASSED!

================================
Checking that push_task() works:
================================
Checking that push_task() works for a function with no arguments or return value...
-> PASSED!
Checking that push_task() works for a function with one argument and no return value...
-> PASSED!
Checking that push_task() works for a function with two arguments and no return value...
-> PASSED!

=============================
Checking that submit() works:
=============================
Checking that submit() works for a function with no arguments or return value...
-> PASSED!
Checking that submit() works for a function with one argument and no return value...
-> PASSED!
Checking that submit() works for a function with two arguments and no return value...
-> PASSED!
Checking that submit() works for a function with no arguments and a return value...
-> PASSED!
Checking that submit() works for a function with one argument and a return value...
-> PASSED!
Checking that submit() works for a function with two arguments and a return value...
-> PASSED!

================================================
Checking that submitting member functions works:
================================================
Checking that push_task() works for a member function with no arguments or return value...
-> PASSED!
Checking that push_task() works for a member function with one argument and no return value...
-> PASSED!
Checking that submit() works for a member function with no arguments or return value...
-> PASSED!
Checking that submit() works for a member function with one argument and no return value...
-> PASSED!
Checking that submit() works for a member function with no arguments and a return value...
-> PASSED!
Checking that submit() works for a member function with one argument and a return value...
-> PASSED!

======================================================================
Checking that submitting member functions from within an object works:
======================================================================
Checking that push_task() works within an object for a member function with no arguments or return value...
-> PASSED!
Checking that push_task() works within an object for a member function with one argument and no return value...
-> PASSED!
Checking that submit() works within an object for a member function with no arguments or return value...
-> PASSED!
Checking that submit() works within an object for a member function with one argument and no return value...
-> PASSED!
Checking that submit() works within an object for a member function with no arguments and a return value...
-> PASSED!
Checking that submit() works within an object for a member function with one argument and a return value...
-> PASSED!

=======================================
Checking that wait_for_tasks() works...
=======================================
Waiting for tasks...
-> PASSED!

======================================================
Checking that push_loop() and parallelize_loop() work:
======================================================
Verifying that push_loop() from 917499 to 884861 with 19 tasks modifies all indices...
-> PASSED!
Verifying that push_loop() from 235488 to 296304 with 11 tasks modifies all indices...
-> PASSED!
Verifying that push_loop() from -790296 to -152228 with 21 tasks modifies all indices...
-> PASSED!
Verifying that push_loop() from -937055 to -135942 with 10 tasks modifies all indices...
-> PASSED!
Verifying that push_loop() from 372276 to 486867 with 3 tasks modifies all indices...
-> PASSED!
Verifying that push_loop() from 890415 to -163491 with 5 tasks modifies all indices...
-> PASSED!
Verifying that push_loop() from 637645 to -894687 with 7 tasks modifies all indices...
-> PASSED!
Verifying that push_loop() from 308032 to -254915 with 20 tasks modifies all indices...
-> PASSED!
Verifying that push_loop() from 499518 to 104936 with 17 tasks modifies all indices...
-> PASSED!
Verifying that push_loop() from 19080 to -378567 with 5 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from -298981 to -724834 with 11 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from 232695 to 767243 with 3 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from 177768 to 966097 with 10 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from 474617 to -155690 with 15 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from -733576 to 547977 with 9 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from -723922 to 992233 with 1 task modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from 957397 to 364478 with 5 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from 776948 to 895847 with 3 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from 696779 to 400637 with 17 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from -5265 to 746418 with 23 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from -229724 to -883103 with 9 tasks correctly sums all indices...
Expected: -727098445812, obtained: -727098445812 -> PASSED!
Verifying that parallelize_loop() from 730130 to -370499 with 3 tasks correctly sums all indices...
Expected: 395819207270, obtained: 395819207270 -> PASSED!
Verifying that parallelize_loop() from -493633 to 957239 with 14 tasks correctly sums all indices...
Expected: 672631513560, obtained: 672631513560 -> PASSED!
Verifying that parallelize_loop() from -455240 to -60850 with 14 tasks correctly sums all indices...
Expected: -203541129490, obtained: -203541129490 -> PASSED!
Verifying that parallelize_loop() from -287333 to 298991 with 9 tasks correctly sums all indices...
Expected: 6834778868, obtained: 6834778868 -> PASSED!
Verifying that parallelize_loop() from 62326 to -392718 with 18 tasks correctly sums all indices...
Expected: -150343352292, obtained: -150343352292 -> PASSED!
Verifying that parallelize_loop() from 663186 to 380865 with 23 tasks correctly sums all indices...
Expected: 294757240050, obtained: 294757240050 -> PASSED!
Verifying that parallelize_loop() from 609125 to -43020 with 1 task correctly sums all indices...
Expected: 369181893080, obtained: 369181893080 -> PASSED!
Verifying that parallelize_loop() from 465469 to 112037 with 4 tasks correctly sums all indices...
Expected: 204108747160, obtained: 204108747160 -> PASSED!
Verifying that parallelize_loop() from 690574 to 113023 with 15 tasks correctly sums all indices...
Expected: 464117673396, obtained: 464117673396 -> PASSED!
Verifying that parallelize_loop() with identical start and end indices does nothing...
-> PASSED!
Trying parallelize_loop() with start and end indices of different types:
Verifying that parallelize_loop() from 894645 to 908567 with 9 tasks modifies all indices...
-> PASSED!
Trying the overloads for push_loop() and parallelize_loop() for the case where the first index is equal to 0:
Verifying that push_loop() from 0 to 949967 with 10 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from 0 to 241018 with 3 tasks modifies all indices...
-> PASSED!
Verifying that parallelize_loop() from 0 to 574984 with 19 tasks correctly sums all indices...
Expected: 330606025272, obtained: 330606025272 -> PASSED!

====================================
Checking that task monitoring works:
====================================
Resetting pool to 4 threads.
Submitting 12 tasks.
After submission, should have: 12 tasks total, 4 tasks running, 8 tasks queued...
Result: 12 tasks total, 4 tasks running, 8 tasks queued -> PASSED!
Task 2 released.
Task 0 released.
Task 1 released.
Task 3 released.
After releasing 4 tasks, should have: 8 tasks total, 4 tasks running, 4 tasks queued...
Result: 8 tasks total, 4 tasks running, 4 tasks queued -> PASSED!
Task 4 released.
Task 7 released.
Task 6 released.
Task 5 released.
After releasing 4 more tasks, should have: 4 tasks total, 4 tasks running, 0 tasks queued...
Result: 4 tasks total, 4 tasks running, 0 tasks queued -> PASSED!
Task 11 released.
Task 10 released.
Task 9 released.
Task 8 released.
After releasing the final 4 tasks, should have: 0 tasks total, 0 tasks running, 0 tasks queued...
Result: 0 tasks total, 0 tasks running, 0 tasks queued -> PASSED!
Resetting pool to 24 threads.

============================
Checking that pausing works:
============================
Resetting pool to 4 threads.
Checking that the pool correctly reports that it is not paused.
-> PASSED!
Pausing pool.
Checking that the pool correctly reports that it is paused.
-> PASSED!
Submitting 12 tasks, each one waiting for 200ms.
Immediately after submission, should have: 12 tasks total, 0 tasks running, 12 tasks queued...
Result: 12 tasks total, 0 tasks running, 12 tasks queued -> PASSED!
300ms later, should still have: 12 tasks total, 0 tasks running, 12 tasks queued...
Result: 12 tasks total, 0 tasks running, 12 tasks queued -> PASSED!
Unpausing pool.
Checking that the pool correctly reports that it is not paused.
-> PASSED!
Task 1 done.
Task 2 done.
Task 0 done.
Task 3 done.
300ms later, should have: 8 tasks total, 4 tasks running, 4 tasks queued...
Result: 8 tasks total, 4 tasks running, 4 tasks queued -> PASSED!
Pausing pool and using wait_for_tasks() to wait for the running tasks.
Task 7 done.
Task 5 done.
Task 6 done.
Task 4 done.
After waiting, should have: 4 tasks total, 0 tasks running, 4 tasks queued...
Result: 4 tasks total, 0 tasks running, 4 tasks queued -> PASSED!
200ms later, should still have: 4 tasks total, 0 tasks running, 4 tasks queued...
Result: 4 tasks total, 0 tasks running, 4 tasks queued -> PASSED!
Unpausing pool and using wait_for_tasks() to wait for all tasks.
Task 8 done.
Task 9 done.
Task 10 done.
Task 11 done.
After waiting, should have: 0 tasks total, 0 tasks running, 0 tasks queued...
Result: 0 tasks total, 0 tasks running, 0 tasks queued -> PASSED!
Resetting pool to 24 threads.

=======================================
Checking that exception handling works:
=======================================
Checking that exceptions are forwarded correctly by submit()...
Throwing exception...
-> PASSED!
Checking that exceptions are forwarded correctly by BS::multi_future...
Throwing exception...
Throwing exception...
-> PASSED!

============================================================
Testing that vector operations produce the expected results:
============================================================
Adding two vectors with 77579 elements using 9 tasks...
-> PASSED!
Adding two vectors with 925926 elements using 2 tasks...
-> PASSED!
Adding two vectors with 367682 elements using 22 tasks...
-> PASSED!
Adding two vectors with 28482 elements using 2 tasks...
-> PASSED!
Adding two vectors with 486607 elements using 19 tasks...
-> PASSED!
Adding two vectors with 688249 elements using 10 tasks...
-> PASSED!
Adding two vectors with 473738 elements using 18 tasks...
-> PASSED!
Adding two vectors with 743021 elements using 12 tasks...
-> PASSED!
Adding two vectors with 209804 elements using 11 tasks...
-> PASSED!
Adding two vectors with 635671 elements using 23 tasks...
-> PASSED!

++++++++++++++++++++++++++++++
SUCCESS: Passed all 88 checks!
++++++++++++++++++++++++++++++

Performance tests

If all checks passed, BS_thread_pool_test.cpp will perform simple benchmarks by filling a specific number of vectors of fixed size with random values. The program decides how many vectors to use by testing how many are needed to reach a target duration in the single-threaded test. This ensures that the test takes approximately the same amount of time on different systems, and is thus more consistent and portable.

Once the required number of vectors has been determined, the program will test the performance of several multi-threaded tests, dividing the total number of vectors into different numbers of tasks, compare them to the performance of the single-threaded test, and indicate the maximum speedup obtained.

Please note that these benchmarks are only intended to demonstrate that the package can provide a significant speedup, and it is highly recommended to perform your own benchmarks with your specific system, compiler, and code.

Here we will present the results of the performance test running on a high-end desktop computer equipped with a 12-core / 24-thread AMD Ryzen 9 3900X CPU at 3.8 GHz and 32 GB of DDR4 RAM at 3600 MHz, compiled using MSVC v19.32.31332 on Windows 11 build 22000.795 with the /O2 compiler flag. The output was as follows:

======================
Performing benchmarks:
======================
Using 24 threads.
Each test will be repeated 20 times to collect reliable statistics.
Determining the number and size of vectors to generate in order to achieve an approximate mean execution time of 50 ms with 24 tasks...
Generating 3312 vectors with 4096 elements each:
Single-threaded, mean execution time was  542.2 ms with standard deviation  5.8 ms.
With    6 tasks, mean execution time was   95.2 ms with standard deviation  1.7 ms.
With   12 tasks, mean execution time was   49.6 ms with standard deviation  0.7 ms.
With   24 tasks, mean execution time was   29.0 ms with standard deviation  2.9 ms.
With   48 tasks, mean execution time was   33.2 ms with standard deviation  4.3 ms.
With   96 tasks, mean execution time was   35.5 ms with standard deviation  1.9 ms.
Maximum speedup obtained by multithreading vs. single-threading: 18.7x, using 24 tasks.

+++++++++++++++++++++++++++++++++++++++
Thread pool performance test completed!
+++++++++++++++++++++++++++++++++++++++

This CPU has 12 physical cores, with each core providing two separate logical cores via hyperthreading, for a total of 24 threads. Without hyperthreading, we would expect a maximum theoretical speedup of 12x. With hyperthreading, one might naively expect to achieve up to a 24x speedup, but this is in fact impossible, as both logical cores share the same physical core's resources. However, generally we would expect an estimated 30% additional speedup from hyperthreading, which amounts to around 15.6x in this case. In our performance test, we see a speedup of 18.7x, saturating and even surpassing this estimated theoretical upper bound.

The light version of the package

This package started out as a very lightweight thread pool, but over time has expanded to include many additional features and helper classes. Therefore, I have decided to bundle a light version of the thread pool in a separate and stand-alone header file, BS_thread_pool_light.hpp, which is about half the size of the full package.

This file does not contain any of the helper classes, only a new BS::thread_pool_light class, which is a minimal thread pool with only the 5 most basic member functions:

  • get_thread_count()
  • push_loop()
  • push_task()
  • submit()
  • wait_for_tasks()

A separate test program BS_thread_pool_light_test.cpp tests only the features of the lightweight BS::thread_pool_light class. In the spirit of minimalism, it does not generate a log file and does not do any benchmarks.

To be perfectly clear, each header file is 100% stand-alone. If you wish to use the full package, you only need BS_thread_pool.hpp, and if you wish to use the light version, you only need BS_thread_pool_light.hpp. Only a single header file needs to be included in your project.

About the project

Issue and pull request policy

This package is under continuous and active development. If you encounter any bugs, or if you would like to request any additional features, please feel free to open a new issue on GitHub and I will look into it as soon as I can.

Contributions are always welcome. However, I release my projects in cumulative updates after editing and testing them locally on my system, so my policy is not to accept any pull requests. If you open a pull request, and I decide to incorporate your suggestion into the project, I will first modify your code to comply with the project's coding conventions (formatting, syntax, naming, comments, programming practices, etc.), and perform some tests to ensure that the change doesn't break anything. I will then merge it into the next release of the project, possibly together with some other changes. The new release will also include a note in CHANGELOG.md with a link to your pull request, and modifications to the documentation in README.md as needed.

Acknowledgements

Many GitHub users have helped improve this project, directly or indirectly, via issues, pull requests, comments, and/or personal correspondence. Please see CHANGELOG.md for links to specific issues and pull requests that have been the most helpful. Thank you all for your contribution! :)

Starring the repository

If you found this project useful, please consider starring it on GitHub! This allows me to see how many people are using my code, and motivates me to keep working to improve it.

Copyright and citing

Copyright (c) 2022 Barak Shoshany. Licensed under the MIT license.

If you use the library in software of any kind, please provide a link to the GitHub repository in the source code and documentation.

If you use this library in published research, please cite it as follows:

You can use the following BibTeX entry:

@article{Shoshany2021_ThreadPool,
    archiveprefix = {arXiv},
    author        = {Barak Shoshany},
    doi           = {10.5281/zenodo.4742687},
    eid           = {arXiv:2105.00613},
    eprint        = {2105.00613},
    journal       = {arXiv e-prints},
    keywords      = {Computer Science - Distributed, Parallel, and Cluster Computing, D.1.3, D.1.5},
    month         = {May},
    primaryclass  = {cs.DC},
    title         = {{A C++17 Thread Pool for High-Performance Scientific Computing}},
    year          = {2021}
}

Please note that the companion paper on arXiv is updated infrequently. The paper is intended to facilitate discovery of the package by scientists who may find it useful for scientific computing purposes and to allow citing the package in scientific research, but most users should read the README.md file on the GitHub repository instead, as it is guaranteed to always be up to date.

About

BS::thread_pool: a fast, lightweight, and easy-to-use C++17 thread pool library

License:MIT License


Languages

Language:C++ 100.0%