A framework for automating workload execution and measurement collection on ARM devices.
Clone or download
scojac01 and marcbonnici Speedometer: Extending teardown function
Some devices throw errors if too many browser tabs are open. Have
added a method to close down the tabs in the teardown function.
Latest commit ccaca3d Sep 11, 2018

README.rst

Workload Automation

Workload Automation (WA) is a framework for executing workloads and collecting measurements on Android and Linux devices. WA includes automation for nearly 40 workloads and supports some common instrumentation (ftrace, hwmon) along with a number of output formats.

WA is designed primarily as a developer tool/framework to facilitate data driven development by providing a method of collecting measurements from a device in a repeatable way.

WA is highly extensible. Most of the concrete functionality is implemented via plug-ins, and it is easy to write new plug-ins to support new device types, workloads, instruments or output processing.

Requirements

  • Python 2.7 or Python 3
  • Linux (should work on other Unixes, but untested)
  • Latest Android SDK (ANDROID_HOME must be set) for Android devices, or
  • SSH for Linux devices

Installation

To install:

git clone git@github.com:ARM-software/workload-automation.git workload-automation
sudo -H pip install ./workload-automation

Please refer to the installation section in the documentation for more details.

Basic Usage

Please see the Quickstart section of the documentation.

Documentation

You can view pre-built HTML documentation here.

Documentation in reStructuredText format may be found under doc/source. To compile it into cross-linked HTML, make sure you have Sphinx installed, and then

cd doc
make html

License

Workload Automation is distributed under Apache v2.0 License. Workload automation includes binaries distributed under different licenses (see LICENSE files in specific directories).

Feedback, Contributions and Support

  • Please use the GitHub Issue Tracker associated with this repository for feedback.
  • ARM licensees may contact ARM directly via their partner managers.
  • We welcome code contributions via GitHub Pull requests. Please see "Contributing Code" section of the documentation for details.