In order to run the optimizer, provide a data file similar to example_data.csv file.
The first row contains the classes of the jobs (e.g. synthesis, placement, routing, sta).
They are encoded starting from 0.
The second row represents the runtime it took on a specific machine configuration (e.g. 1, 2, 4, 8 vCPUs). Note that the order matters as the optimization uses the inices of these runtimes to relate them to a machine congiguration.
The third row represents the hourly cost for running that configuration for that stage.
To run:
python cloud_optimizer.py <budget> -data <data_file>For example:
python cloud_optimizer.py 3000 -data example_data.csv
This will output the total runtime, the total cost, and the recommended machine configuration (index) for each class.