AnsweredAssumed Answered

Can you describe the source code relationship description between your components and others in the ecosystem?

Question asked by jacques on Jun 30, 2011
Latest reply on Jun 30, 2011 by aaron
As a user of other Hadoop products, I'm finding it difficult understanding the relationship between your components and others out in the ecosystem.  My understanding:

 - CLDB + FileServer: completely replaces NameNode + DataNode.  No shared code, new code written in C++
 - Map Reduce: Based off a 20.x release but heavily modified to perform better
 - HBase: Mostly the same as a 90.x release (??) but smallish bug fixes
 - Hive: (???)
 - Flume: (0.93 or 0.94 based?)
 - Sqoop, Mahout, pig, etc ???

I understand that your releases are separate but it would be helpful to fully understand provenance.  Cloudera does a good job of this with their list of additionally incorporated bug fixes for each release.  Since we'll continue to rely at least partly on the community at large if we choose to use your stack, we need to be able to relate to the communities' version numbers and fixes.

Outcomes