Installing and deploying Omniperf#

Omniperf consists of two installation components.

  • Omniperf core installation (client-side)

    • Provides the core application profiling capability.

    • Allows the collection of performance counters, filtering by hardware block, dispatch, kernel, and more.

    • Provides a CLI-based analysis mode.

    • Provides a standalone web interface for importing analysis metrics.

  • Grafana server for Omniperf (server-side) (optional)

    • Hosts the MongoDB backend and Grafana instance.

    • Is packaged in a Docker container for easy setup.

Determine what you need to install based on how you would like to interact with Omniperf. See the following decision tree to help determine what installation is right for you.

Decision tree for installing and deploying Omniperf

Core installation#

The core Omniperf application requires the following basic software dependencies. As of ROCm 6.2, the core Omniperf is included with your ROCm installation.

  • Python >= 3.8

  • CMake >= 3.19

  • ROCm >= 5.7.1

Note

Omniperf will use the first version of Python3 found in your system’s PATH. If the default version of Python3 is older than 3.8, you may need to update your system’s PATH to point to a newer version of Python3.

Omniperf depends on a number of Python packages documented in the top-level requirements.txt file. Install these before configuring Omniperf.

Tip

If looking to build Omniperf as a developer, consider these additional requirements.

docs/sphinx/requirements.txt

Python packages required to build this documentation from source.

requirements-test.txt

Python packages required to run Omniperf’s CI suite using PyTest.

The recommended procedure for Omniperf usage is to install into a shared file system so that multiple users can access the final installation. The following steps illustrate how to install the necessary Python dependencies using pip and Omniperf into a shared location controlled by the INSTALL_DIR environment variable.

Tip

To always run Omniperf with a particular version of python, you can create a bash alias. For example, to run Omniperf with Python 3.10, you can run the following command:

alias omniperf-mypython="/usr/bin/python3.10 /opt/rocm/bin/omniperf"

Configuration variables#

The following installation example leverages several CMake project variables defined as follows.

CMake variable

Description

CMAKE_INSTALL_PREFIX

Controls the install path for Omniperf files.

PYTHON_DEPS

Specifies an optional path to resolve Python package dependencies.

MOD_INSTALL_PATH

Specifies an optional path for separate Omniperf modulefile installation.

Install from source#

  1. A typical install begins by downloading the latest release tarball available from ROCm/omniperf. From there, untar and navigate into the top-level directory.

    tar xfz omniperf-v2.0.1.tar.gz
    cd omniperf-v2.0.1
    
  2. Next, install Python dependencies and complete the Omniperf configuration and install process.

    # define top-level install path
    export INSTALL_DIR=<your-top-level-desired-install-path>
    
    # install python deps
    python3 -m pip install -t ${INSTALL_DIR}/python-libs -r requirements.txt
    
    # configure Omniperf for shared install
    mkdir build
    cd build
    cmake -DCMAKE_INSTALL_PREFIX=${INSTALL_DIR}/2.0.1 \
            -DPYTHON_DEPS=${INSTALL_DIR}/python-libs \
            -DMOD_INSTALL_PATH=${INSTALL_DIR}/modulefiles ..
    
    # install
    make install
    

    Tip

    You might need to sudo the final installation step if you don’t have write access for the chosen installation path.

  3. Upon successful installation, your top-level installation directory should look like this.

    $ ls $INSTALL_DIR
    modulefiles  2.0.1  python-libs
    

Execution using modulefiles#

The installation process includes the creation of an environment modulefile for use with Lmod. On systems that support Lmod, you can register the Omniperf modulefile directory and setup your environment for execution of Omniperf as follows.

$ module use $INSTALL_DIR/modulefiles
$ module load omniperf
$ which omniperf
/opt/apps/omniperf/2.0.1/bin/omniperf

$ omniperf --version
ROC Profiler:   /opt/rocm-5.1.0/bin/rocprof

omniperf (v2.0.1)

Tip

If you’re relying on an Lmod Python module locally, you may wish to customize the resulting Omniperf modulefile post-installation to include extra module dependencies.

Execution without modulefiles#

To use Omniperf without the companion modulefile, update your PATH settings to enable access to the command line binary. If you installed Python dependencies in a shared location, also update your PYTHONPATH configuration.

export PATH=$INSTALL_DIR/2.0.1/bin:$PATH
export PYTHONPATH=$INSTALL_DIR/python-libs

Install via package manager#

Once ROCm (minimum version 6.2.0) is installed, you can install Omniperf using your operating system’s native package manager using the following commands. See ROCm installation for Linux for guidance on installing the ROCm software stack.

$ sudo apt install omniperf
$ python3 -m pip install -r /opt/rocm/libexec/omniperf/requirements.txt
$ sudo dnf install omniperf
$ python3 -m pip install -r /opt/rocm/libexec/omniperf/requirements.txt
$ sudo zypper install omniperf
$ python3 -m pip install -r /opt/rocm/libexec/omniperf/requirements.txt

ROCProfiler#

Omniperf relies on ROCProfiler’s rocprof binary during the profiling process. Normally, the path to this binary is detected automatically, but you can override the path by the setting the optional ROCPROF environment variable.