Home > Workload Solutions > High Performance Computing > White Papers > Machine Learning Using Red Hat OpenShift Container Platform > Business case
Enterprises are increasing their investments in infrastructure platforms to support Artificial Intelligence (AI) use cases and the computing needs of their data science teams. Machine Learning (ML) and Deep Learning (DL) are AI techniques that have demonstrated success across every industry vertical, including manufacturing, healthcare, retail, and cloud services.
Kubeflow, a Kubernetes-native platform for ML workloads for enterprises, was released as an open-source project in December 2017. Kubeflow is a composable, scalable, portable ML stack that was originally based on Google’s use of TensorFlow on Kubernetes but now includes components and contributions from several sources and organizations. Eventually, most popular ML tools will become a part of Kubeflow. Kubeflow makes it easier to develop, deploy, and manage ML applications. For more information, see Kubeflow: The Machine Learning Toolkit for Kubernetes.
Kubeflow requires a Kubernetes environment such as Google Kubernetes Engine or Red Hat OpenShift. Running Kubeflow on OpenShift offers several advantages in an ML context: