Open Data Hub 0.5.1 Release Guide

What is included?

Open Data Hub 0.5.1 includes many new tools that are essential to a comprehensive AI/ML end-to-end platform. Open Data Hub is a meta-operator that can be installed on Openshift Container Platform 3.11 and 4.x.

The following is a list of tools added to Open Data Hub in this release:

Technology Version Category
JupyterHub 3.0.7 Data science tools
Radanalytics Spark Operator 1.0.5 Operator for managing Spark cluster on OpenShift
Apache Superset 0.34.0 Data Exploration and Visualization Tool

You can review the release notes for components added in the previous v0.5.0 release here

AICoE-JupyterHub

AICoE-JupyterHub now has support for greater customization of the JupyterHub deployment and Spark cluster resources. In this release, we added

  • Support for modifying the JupyterHub server behavior via a custom JupyterHub config
  • Update the kubespawner library to version 0.11.1
  • Support for specifying the resource requests and limits of the cpu/memory allocated to the Jupyter notebook user spark cluster nodes using the default JupyterHub Singleuser profile

Radanalytics Spark Operator

  • Support for specifying the resource requests and limits of the cpu/memory allocated to the Spark cluster master and worker nodes

Apache Superset

  • Resolve issues related to connecting to the Data Catalog Thrift Server