Mailing-List: contact dev-help@hive.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@hive.apache.org
Date: Fri, 18 Jan 2013 18:08:15 +0000 (UTC)
From: "Ashutosh Chauhan (JIRA)" <jira@apache.org>
To: hive-dev@hadoop.apache.org
Message-ID: <JIRA.12509546.1307515110478.159581.1358532495551@arcas>
In-Reply-To: <JIRA.12509546.1307515110478@arcas>
References: <JIRA.12509546.1307515110478@arcas>
Subject: [jira] [Commented] (HIVE-2206) add a new optimizer for query
 correlation discovery and optimization
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable


    [ https://issues.apache.org/jira/browse/HIVE-2206?page=3Dcom.atlassian.=
jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D13557=
411#comment-13557411 ]=20

Ashutosh Chauhan commented on HIVE-2206:
----------------------------------------

I am having second thoughts on cloning. Cloning graphs (like query plan) or=
 dense structures (like ParseContext) is fraught with perils. Its likely th=
at cloning will require new code and arguably have hard to detect bugs, sin=
ce we need to track down every single pointer and clone all the way through=
. I think to avoid such issues and for simplicity, we can drop the cloning =
idea. The feature is anyway behind the config option which is off default, =
so query-plan will be modified only for the users who turn the flag on.=20
Yin, if you have addressed my other comments, can you update the patch on R=
B and upload here on jira, I will take another look at it.
               =20
> add a new optimizer for query correlation discovery and optimization
> --------------------------------------------------------------------
>
>                 Key: HIVE-2206
>                 URL: https://issues.apache.org/jira/browse/HIVE-2206
>             Project: Hive
>          Issue Type: New Feature
>          Components: Query Processor
>    Affects Versions: 0.10.0
>            Reporter: He Yongqiang
>            Assignee: Yin Huai
>         Attachments: HIVE-2206.10-r1384442.patch.txt, HIVE-2206.11-r13850=
84.patch.txt, HIVE-2206.12-r1386996.patch.txt, HIVE-2206.13-r1389072.patch.=
txt, HIVE-2206.14-r1389704.patch.txt, HIVE-2206.15-r1392491.patch.txt, HIVE=
-2206.16-r1399936.patch.txt, HIVE-2206.17-r1404933.patch.txt, HIVE-2206.18-=
r1407720.patch.txt, HIVE-2206.19-r1410581.patch.txt, HIVE-2206.1.patch.txt,=
 HIVE-2206.20-r1434012.patch.txt, HIVE-2206.2.patch.txt, HIVE-2206.3.patch.=
txt, HIVE-2206.4.patch.txt, HIVE-2206.5-1.patch.txt, HIVE-2206.5.patch.txt,=
 HIVE-2206.6.patch.txt, HIVE-2206.7.patch.txt, HIVE-2206.8.r1224646.patch.t=
xt, HIVE-2206.8-r1237253.patch.txt, testQueries.2.q, YSmartPatchForHive.pat=
ch
>
>
> This issue proposes a new logical optimizer called Correlation Optimizer,=
 which is used to merge correlated MapReduce jobs (MR jobs) into a single M=
R job. The idea is based on YSmart (http://ysmart.cse.ohio-state.edu/).The =
paper and slides of YSmart are linked at the bottom.
> Since Hive translates queries in a sentence by sentence fashion, for ever=
y operation which may need to shuffle the data (e.g. join and aggregation o=
perations), Hive will generate a MapReduce job for that operation. However,=
 for those operations which may need to shuffle the data, they may involve =
correlations explained below and thus can be executed in a single MR job.
> # Input Correlation: Multiple MR jobs have input correlation (IC) if thei=
r input relation sets are not disjoint;
> # Transit Correlation: Multiple MR jobs have transit correlation (TC) if =
they have not only input correlation, but also the same partition key;
> # Job Flow Correlation: An MR has job =EF=AC=82ow correlation (JFC) with =
one of its child nodes if it has the same partition key as that child node.
> The current implementation of correlation optimizer only detect correlati=
ons among MR jobs for reduce-side join operators and reduce-side aggregatio=
n operators (not map only aggregation). A query will be optimized if it sat=
isfies following conditions.
> # There exists a MR job for reduce-side join operator or reduce side aggr=
egation operator which have JFC with all of its parents MR jobs (TCs will b=
e also exploited if JFC exists);
> # All input tables of those correlated MR job are original input tables (=
not intermediate tables generated by sub-queries); and=20
> # No self join is involved in those correlated MR jobs.
> Correlation optimizer is implemented as a logical optimizer. The main rea=
sons are that it only needs to manipulate the query plan tree and it can le=
verage the existing component on generating MR jobs.
> Current implementation can serve as a framework for correlation related o=
ptimizations. I think that it is better than adding individual optimizers.=
=20
> There are several work that can be done in future to improve this optimiz=
er. Here are three examples.
> # Support queries only involve TC;
> # Support queries in which input tables of correlated MR jobs involves in=
termediate tables; and=20
> # Optimize queries involving self join.=20
> References:
> Paper and presentation of YSmart.
> Paper: http://www.cse.ohio-state.edu/hpcs/WWW/HTML/publications/papers/TR=
-11-7.pdf
> Slides: http://sdrv.ms/UpwJJc

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrato=
rs
For more information on JIRA, see: http://www.atlassian.com/software/jira