RE: How does Diyotta allow working with two Hadoop clusters of different versions simultaneously without any library related conflicts?
Diyotta does allow using multiple Hadoop clusters of different versions simultaneously. Diyotta data point seeks the version information of the Hadoop distribution along with other connection information. For each combination of version and distribution, a separate agent service is spun off from the parent agent with its own dependencies. This avoids any conflict between dependencies of different versions.