Minizinc models of a power-aware task placement onto a heterogenous platform (.e.g, big-little, gpu, fpga).

$ sudo snap install minizinc
$ sudo apt-get install python3-venv
$ git clone https://github.com/amamory/power-optim.git
$ cd power-optim
$ python3 -m venv env
$ source env/bin/activate
$ pip install --editable .
$ pip install -r requirements.txt

Browse the directories to try each model. The suggested order in terms of complexity:

• makespan: jobs without precedence constraint;
• makespan-dag: similar to the previous one but jobs have precedence constraints represented by a DAG;
• deadline-dag: almost equal to the previous one, but it includes an additional deadline constraint;
• deadline-power: while the previous one minimizes the makespan, this one minizes the power as long as the deadline is respected;
• affinity: this model is built on top of the previous one, but adding affinity constraint;
• pu-utilization: this model is built on top of the previous one, but including task period and processing unit utilization constraint. TODO.

• Alexandre Amory (January 2022), Real-Time Systems Laboratory (ReTiS Lab), Scuola Superiore Sant'Anna (SSSA), Pisa, Italy.

This tool has been developed in the context of the AMPERE project. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871669.