README.rst

Demo for using COBalD

Toy demo showing how the cobald daemon predicts resources required for unknown demand. It implements a dummy pool which tracks your CPU usage, and exposes its utilisation compared to a faked demand. This is visualised as a simple ASCII art on your terminal:

[-----------------------------------^----------------------------v-----------------------------------]
[------------------------------------^-----v---------------------------------------------------------]
[------------------------------v-----^---------------------------------------------------------------]
[-------------------------------------^------------v-------------------------------------------------]
[----------------v--------------------^--------------------------------------------------------------]
[-----------------v------------------^---------------------------------------------------------------]
[-------------v----------------------^---------------------------------------------------------------]
              | cpu usage            | target demand

Usage

You can pull in the entire demo and its dependencies using git and pip. Install it in a venv to isolate the demo:

python3 -m venv /tmp/cobalddemo/venv/ --clear
cd /tmp/cobalddemo/
git clone https://github.com/MaineKuehn/cobald_demo.git
source venv/bin/activate
python3 -m pip install --editable cobald_demo

After installation, you can run the demo with one of the two example configuration files:

python3 -m cobald.daemon cobald_demo/demo_config.yaml
python3 -m cobald.daemon cobald_demo/demo_config.py

Looking around

If you just want to use cobald with existing components, check out the configuration files. The demo_config.yaml uses a simple but limited format, while the demo_config.py provides a complex but extensible approach.

If you want to write your own resource pool, check out the cobald_demo/cpu_pool.py implementation. While it is a mockup, it shows how little is needed to hook into the cobald machinery.