# C++ Examples ## Tutorials Check out our [tutorials](../tutorials) for step-by-step guides on using MSCCL++. (mscclpp-test)= ## mscclpp-test *NOTE: mscclpp-test is NOT a performance benchmark. If you want to get the latest performance numbers, please use the Python benchmark or the NCCL APIs instead.* mscclpp-test is a set of C++ implementation examples. It requires MPI on the system, and the path should be provided via `MPI_HOME` environment variable to the CMake build system. ```bash $ MPI_HOME=/path/to/mpi cmake -DCMAKE_BUILD_TYPE=Release .. $ make -j allgather_test_perf allreduce_test_perf ``` For example, the following command runs the `allreduce5` algorithm with 8 GPUs starting from 3MB to 48MB messages, by doubling the message size in between. You can try different algorithms by changing the `-k 5` option to another value (e.g., `-k 3` runs `allreduce3`). Check all algorithms from the code: [allreduce_test.cu](https://github.com/microsoft/mscclpp/blob/main/test/mscclpp-test/allreduce_test.cu) and [allgather_test.cu](https://github.com/microsoft/mscclpp/blob/main/test/mscclpp-test/allgather_test.cu). ```bash $ mpirun --bind-to numa -np 8 ./test/mscclpp-test/allreduce_test_perf -b 3m -e 48m -G 100 -n 100 -w 20 -f 2 -k 5 ``` *NOTE: a few algorithms set a condition on the total data size, such as to be a multiple of 3. If the condition is unmet, the command will throw a regarding error.* Check the help message for more details. ```bash $ ./test/mscclpp-test/allreduce_test_perf --help USAGE: allreduce_test_perf [-b,--minbytes ] [-e,--maxbytes ] [-i,--stepbytes ] [-f,--stepfactor ] [-n,--iters ] [-w,--warmup_iters ] [-c,--check <0/1>] [-T,--timeout