Ambari is an open source project, but HDP is no longer available for download, so how does Ambari manage components such as Apache Hadoop, Hive, and Spark? Can Ambari directly manage existing Hadoop clusters?