namespace tf {

/** @page BenchmarkTaskflow Benchmark Taskflow

@tableofcontents

@section CompileAndRunBenchmarks Compile and Run Benchmarks

To build the benchmark code,
enable the CMake option @c TF_BUILD_BENCHMARKS to @c ON as follows:

@code{.shell-session}
# under /taskflow/build
~$ cmake ../ -DTF_BUILD_BENCHMARKS=ON
~$ make
@endcode

After you successfully build the benchmark code, you can find all benchmark
instances in the @c benchmarks/ folder.
You can run the executable of each instance in the corresponding folder.

@code{.shell-session}
~$ cd benchmarks & ls
bench_black_scholes bench_binary_tree bench_graph_traversal ...
~$ ./bench_graph_traversal
|V|+|E|     Runtime
      2       0.197
    842       0.198
   3284       0.488
   7288       0.774
    ...         ...
    ...         ...
 619802      75.135
 664771      77.436
 711200      83.957
@endcode

You can display the help message by giving the option @c --help.

@code{.shell-session}
~$ ./bench_graph_traversal --help
Graph Traversal
Usage: ./bench_graph_traversal [OPTIONS]

Options:
  -h,--help                   Print this help message and exit
  -t,--num_threads UINT       number of threads (default=1)
  -r,--num_rounds UINT        number of rounds (default=1)
  -m,--model TEXT             model name tbb|omp|tf (default=tf)
@endcode

We currently implement the following instances that are commonly used by
the parallel computing community to evaluate the system performance.

| Instance | Description |
| :-:      | :-:         |
| bench_binary_tree | traverses a complete binary tree |
| bench_black_scholes | computes option pricing with Black-Shcoles Models |
| bench_graph_traversal | traverses a randomly generated direct acyclic graph |
| bench_linear_chain    | traverses a linear chain of tasks |
| bench_mandelbrot      | exploits imbalanced workloads in a Mandelbrot set |
| bench_matrix_multiplication | multiplies two 2D matrices |
| bench_mnist | trains a neural network-based image classifier on the MNIST dataset |
| bench_parallel_sort | sorts a range of items |
| bench_reduce_sum | sums a range of items using reduction |
| bench_wavefront | propagates computations in a 2D grid |
| bench_linear_pipeline | pipeline scheduling on a linear chain of pipes |
| bench_graph_pipeline | pipeline scheduling on a graph of pipes |


@section ConfigureRunOptions Configure Run Options

We implement consistent options for each benchmark instance. 
Common options are:

| option | value | function |
| :-:    | :-:   | :-:      |
| @c -h  | none  | display the help message |
| @c -t  | integer | configure the number of threads to run |
| @c -r  | integer | configure the number of rounds to run |
| @c -m  | string | configure the baseline models to run, tbb, omp, or tf |

You can configure the benchmarking environment by giving different options.

@subsection SpecifyTheRunModel Specify the Run Model

In addition to a %Taskflow-based implementation for each benchmark instance,
we have implemented two baseline models using the state-of-the-art parallel
programming libraries, @OpenMP and @TBB,
to measure and evaluate the performance of %Taskflow.
You can select different implementations by passing the option @c -m.

@code{.shell-session}
~$ ./bench_graph_traversal -m tf   # run the Taskflow implementation (default)
~$ ./bench_graph_traversal -m tbb  # run the TBB implementation
~$ ./bench_graph_traversal -m omp  # run the OpenMP implementation
@endcode

@subsection SpecifyTheNumberOfThreads Specify the Number of Threads

You can configure the number of threads to run a benchmark instance
by passing the option @c -t. 
The default value is one.

@code{.shell-session}
# run the Taskflow implementation using 4 threads
~$ ./bench_graph_traversal -m tf -t 4
@endcode

Depending on your environment, you may need to use @c taskset to set the CPU
affinity of the running process.
This allows the OS scheduler to keep process on the same CPU(s) as long as
practical for performance reason.

@code{.shell-session}
# affine the process to 4 CPUs, CPU 0, CPU 1, CPU 2, and CPU 3
~$ taskset -c 0-3 bench_graph_traversal -t 4  
@endcode

@subsection SpecifyTheNumberOfRounds Specify the Number of Rounds

Each benchmark instance evaluates the runtime of the implementation 
at different problem sizes.
Each problem size corresponds to one iteration.
You can configure the number of rounds per iteration to average the runtime.

@code{.shell-session}
# measure the runtime in an average of 10 runs
~$ ./bench_graph_traversal -r 10
|V|+|E|     Runtime
      2       0.109   # the runtime value 0.109 is an average of 10 runs
    842       0.298
    ...         ...
 619802      73.135
 664771      74.436
@endcode

*/


}