Google announces Kubernetes Operator for Apache Spark

The beta release of "Spark Operator" allows native execution of Spark applications on Kubernetes clusters -- no Hadoop or Mesos required.