Question from the Azure Machine Learning Studio test

What technology is used for model deployment when creating a "Compute clusters" type compute resource in the Azure Machine Learning Studio environment?