Mailing-List: contact dev-help@pig.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@pig.apache.org
Date: Thu, 11 Sep 2014 13:32:34 +0000 (UTC)
From: "Quang-Nhat HOANG-XUAN (JIRA)" <jira@apache.org>
To: pig-dev@hadoop.apache.org
Message-ID: <JIRA.12728811.1406020277000.9071.1410442354513@Atlassian.JIRA>
In-Reply-To: <JIRA.12728811.1406020277000@Atlassian.JIRA>
References: <JIRA.12728811.1406020277000@Atlassian.JIRA>
 <JIRA.12728811.1406020277758@arcas>
Subject: [jira] [Updated] (PIG-4066) An optimization for ROLLUP operation in
 Pig
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable


     [ https://issues.apache.org/jira/browse/PIG-4066?page=3Dcom.atlassian.=
jira.plugin.system.issuetabpanels:all-tabpanel ]

Quang-Nhat HOANG-XUAN updated PIG-4066:
---------------------------------------
    Attachment: TechnicalNotes.2.pdf

> An optimization for ROLLUP operation in Pig
> -------------------------------------------
>
>                 Key: PIG-4066
>                 URL: https://issues.apache.org/jira/browse/PIG-4066
>             Project: Pig
>          Issue Type: Improvement
>            Reporter: Quang-Nhat HOANG-XUAN
>            Assignee: Quang-Nhat HOANG-XUAN
>              Labels: hybrid-irg, optimization, rollup
>         Attachments: Current Rollup vs Our Rollup.jpg, PIG-4066.2.patch, =
PIG-4066.patch, TechnicalNotes.2.pdf, TechnicalNotes.pdf, UserGuide.pdf
>
>
> This patch aims at addressing the current limitation of the ROLLUP operat=
or in PIG: most of the work is done in the Map phase of the underlying MapR=
educe job to generate all possible intermediate keys that the reducer use t=
o aggregate and produce the ROLLUP output. Based on our previous work: =E2=
=80=9CDuy-Hung Phan, Matteo Dell=E2=80=99Amico, Pietro Michiardi: On the de=
sign space of MapReduce ROLLUP aggregates=E2=80=9D (http://www.eurecom.fr/e=
n/publication/4212/download/rs-publi-4212_2.pdf), we show that the design s=
pace for a ROLLUP implementation allows for a different approach (in-reduce=
r grouping, IRG), in which less work is done in the Map phase and the group=
ing is done in the Reduce phase. This patch presents the most efficient imp=
lementation we designed (Hybrid IRG), which allows defining a parameter to =
balance between parallelism (in the reducers) and communication cost.
> This patch contains the following features:
> 1. The new ROLLUP approach: IRG, Hybrid IRG.
> 2. The PIVOT clause in CUBE operators.
> 3. Test cases.
> The new syntax to use our ROLLUP approach:
> alias =3D CUBE rel BY { CUBE col_ref | ROLLUP col_ref [PIVOT pivot_value]=
} [, { CUBE col_ref | ROLLUP col_ref [PIVOT pivot_value]}...]
> In case there is multiple ROLLUP operator in one CUBE clause, the last RO=
LLUP operator will be executed with our approach (IRG, Hybrid IRG) while th=
e remaining ROLLUP ahead will be executed with the default approach.
> We have already made some experiments for comparison between our ROLLUP i=
mplementation and the current ROLLUP. More information can be found at here=
: http://hxquangnhat.github.io/PIG-ROLLUP-H2IRG/
> Patch can be reviewed at here: https://reviews.apache.org/r/23804/


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)