impala-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dan Hecht (Code Review)" <ger...@cloudera.org>
Subject [Impala-ASF-CR] IMPALA-5784 : Separate planner and user set query options in profile
Date Wed, 23 Aug 2017 00:01:53 GMT
Dan Hecht has posted comments on this change.

Change subject: IMPALA-5784 : Separate planner and user set query options in profile
......................................................................


Patch Set 1:

> > > > > (1 comment)
 > > > > Relying to Dan and Balasz:
 > > > >
 > > > > It's actually hard to determine what gets set where using the
 > > > > QueryOptions map thing we have right now, mostly because of
 > how
 > > > > inconsistent we are with default values. E.g. what if a
 > > property
 > > > is
 > > > > set in the session, then the planner sets it as well, but
 > > happens
 > > > > to set it to the default value. Right now we can't even
 > > identify
 > > > > that as a non-default query option.
 > > > >
 > > >
 > > > I don't follow this. We're defining "non-default" query option
 > to
 > > > mean everything that doesn't match the static query option
 > value,
 > > > right?
 > > >
 > > > If so, why can't we just compare the query option values after
 > > > planning and to the values before planning, and print any
 > > > differences as "planner set" (or whatever name).  And then
 > > compare
 > > > the query options before planning with the static defaults,
 > > remove
 > > > anything we've already printed as "planner set" and print those
 > > > remaining differences as "manually set".
 > > >
 > > > or are you saying that doing that is tricky due to the
 > > > datastructures we currently have?
 > >
 > > Yes, I'm saying the data structures we have right now don't lend
 > > themselves to what you're describing. Its certainly possible, but
 > > it'd be awkward given how it currently works and I think it'd be
 > > kind of a pain to test and make sure we don't break different
 > cases
 > > (e.g. values set in the session/query x modified or not by the
 > > planner x with/without a static default). If you think it's worth
 > > it in terms of usability, then Bikram can keep exploring that
 > path.
 > 
 > The code in query-options.cc/h is pretty hairy, e.g.
 > https://github.com/apache/incubator-impala/blob/master/be/src/service/query-options.cc#L82

Okay, let's go with this for now then, but please see Jeszy's naming suggestion.

-- 
To view, visit http://gerrit.cloudera.org:8080/7721
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Ibfc7832963fa0bd278a45c06a5a54e1bf40d8876
Gerrit-PatchSet: 1
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-Owner: Bikramjeet Vig <bikramjeet.vig@cloudera.com>
Gerrit-Reviewer: Balazs Jeszenszky <jeszyb@gmail.com>
Gerrit-Reviewer: Bikramjeet Vig <bikramjeet.vig@cloudera.com>
Gerrit-Reviewer: Dan Hecht <dhecht@cloudera.com>
Gerrit-Reviewer: Matthew Jacobs <mj@cloudera.com>
Gerrit-HasComments: No

Mime
View raw message