End-to-End Machine Learning Using Containerization

Lately, we've been talking a lot about containerization and how Kubernetes and MapR can pair up to enhance the productivity of your data science teams and decrease the time to insights. In this multi-part blog series, I will start with a high-level overview of why Kubernetes and containerization are appealing for data science environments. In a later iteration, I will provide an example of a framework that enables Kubernetized data science on your MapR cluster. Earlier this year, we released the MapR Volume Driver for Kubernetes, which enabled MapR customers to use Kubernetes clusters as extensions of their MapR computing space. This volume plugin provides the ability to mount directories from the MapR global namespace easily to Kubernetes pods, enabling stateful applications to run using your data in place.