MiCADO - Scaling Optimizer with Machine Learning Support

Don't forget to set environment

From project root run source activate optimizer

Test program

From project root run python pure.py

Test flask

From project root run python helloMTA.py

Start program

From project root run python optimizer.py --cfg path/to_config_file

python optimizer.py --cfg config/config.yaml

python optimizer.py --cfg config/config.yaml --host=192.168.0.60

python optimizer.py --cfg config/config.yaml --host 0.0.0.0 --port 5000

Test REST API

GET /optimizer/hello Test REST API. curl -X GET http://193.224.59.115:5000/optimizer/hello

curl -X GET http://193.224.59.115:5000/

POST /optimizer/init Initialize optimizer with the neccessary constants. curl -X POST http://127.0.0.1:5000/optimizer/init --data-binary @test_files/optimizer_constants.yaml

curl -X POST http://193.224.59.115:5000/optimizer/init --data-binary @test_files/optimizer_constants.yaml

POST /optimizer/sample Send a new training sample. curl -X POST http://127.0.0.1:5000/optimizer/sample --data-binary @test_files/metrics_sample_example.yaml

curl -X POST http://193.224.59.115:5000/optimizer/sample --data-binary @test_files/metrics_sample_example.yaml

curl -X POST http://193.224.59.115:5000/optimizer/sample --data-binary @test_files/metrics_sample_example_up.yaml

curl -X POST http://193.224.59.115:5000/optimizer/sample --data-binary @test_files/metrics_sample_example_down.yaml

GET /optimizer/advice Get scaling advice. curl -X GET http://127.0.0.1:5000/optimizer/advice

curl -X GET http://193.224.59.115:5000/optimizer/advice

curl -X GET http://193.224.59.115:5000/optimizer/advice?last=False

GET /optimizer/training_data Download zipped training data that contains both neural network and linear regression data. curl -X GET http://127.0.0.1:5000/optimizer/training_data

TEST CSV

cd csv/csv_to_optimizer source env/csv_to_optimizer/bin/activate python csv_to_optimizer.py