Runtime PMC Meeting Notes 04.12.2018

Meeting agenda:

1. TSC meeting feedback for proposing on whether there is still a need for a reference implementation. General feedback here, were are good on focusing the effort on what we have time for. Ginesh? Reach Out to him and connect him with the meeting time. Reach out to [email protected].

2. Contributors for ODPi Runtime v3 John had sent out an email, any responses? John will be sending out email to others.

3. Scope of ODPi Runtime v3 specification? Take action item, each of us will work with people to figure out what we need to worry about for hadoop v3 vs what we had in the past. Connection big sql with HDP.

Alan wo weeks from now, David has some feedback. Started with referencing the Hadoop v2 spec, everything is still relevant. Saas is starting to do work with Spark as an alternative to MapReduce work. Reference to kerberos. Talked about Ranger? We have talked about including Ranger.

Recommend Spark / Ranger into the spec. Is knox as pressing as ranger?

Robin Nobel, reach out to her on hadoop v3 integration. One thing to help Alan out, feeling to the group on what components do we want to include. Who should we reach out to. Get someone from the hadoop team for v2 to v3. What can we can contain?

4. Using BigTop as Test Framework Test framework for v3 Action item to investigate BigTop

The bigtop tests are setup to build an artifact out of bigtop that only includes the dependencies that it needs.

The bigtop tests artifact that comes out of the bigtop-tests build.gradle is a subset of bigtop 1/2MB https://www.apache.org/dist/bigtop/1.2.0/bigtop-1.2.0-testsuite.zip.

Test output see what we can do there to have a user friendly report.

Attendees:

• John Mertic (LF)
• Tim Thrope (IBM)
• David (Saas)
• Alan Gates (Hortonworks)
• Raj Desai (IBM)