Google introduced new capabilities for managing clusters of Hadoop and Spark.
Google Cloud Dataproc, which is now in beta, is a managed Spark and Hadoop service that leverages open source data tools for batch processing, querying, streaming, and machine learning. The service can be used to create and manage clusters ranging in size from 3 to hundreds of nodes.
Google said its Cloud Dataproc can create Spark and Hadoop clusters in 90 seconds or less, compared to 5 to 30 minutes using on-premises or IaaS providers.
