Return-Path: 
 <issues-return-5152-apmail-hive-issues-archive=hive.apache.org@hive.apache.org>
X-Original-To: apmail-hive-issues-archive@minotaur.apache.org
Delivered-To: apmail-hive-issues-archive@minotaur.apache.org
Received: from mail.apache.org (hermes.apache.org [140.211.11.3])
	by minotaur.apache.org (Postfix) with SMTP id CECBA17987
	for <apmail-hive-issues-archive@minotaur.apache.org>;
 Tue, 21 Apr 2015 13:34:04 +0000 (UTC)
Received: (qmail 17222 invoked by uid 500); 21 Apr 2015 13:33:59 -0000
Delivered-To: apmail-hive-issues-archive@hive.apache.org
Received: (qmail 17096 invoked by uid 500); 21 Apr 2015 13:33:59 -0000
Mailing-List: contact issues-help@hive.apache.org; run by ezmlm
Precedence: bulk
List-Help: <mailto:issues-help@hive.apache.org>
List-Unsubscribe: <mailto:issues-unsubscribe@hive.apache.org>
List-Post: <mailto:issues@hive.apache.org>
List-Id: <issues.hive.apache.org>
Reply-To: dev@hive.apache.org
Delivered-To: mailing list issues@hive.apache.org
Received: (qmail 17078 invoked by uid 99); 21 Apr 2015 13:33:59 -0000
Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28)
    by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 21 Apr 2015 13:33:59 +0000
Date: Tue, 21 Apr 2015 13:33:59 +0000 (UTC)
From: "Gopal V (JIRA)" <jira@apache.org>
To: issues@hive.apache.org
Message-ID: <JIRA.12818812.1428387282000.61826.1429623239667@Atlassian.JIRA>
In-Reply-To: <JIRA.12818812.1428387282000@Atlassian.JIRA>
References: <JIRA.12818812.1428387282000@Atlassian.JIRA>
 <JIRA.12818812.1428387282037@arcas>
Subject: [jira] [Commented] (HIVE-10235) Loop optimization for SIMD in
 ColumnDivideColumn.txt
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394


    [ https://issues.apache.org/jira/browse/HIVE-10235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14504949#comment-14504949 ] 

Gopal V commented on HIVE-10235:
--------------------------------

[~chengxiang li]: Patch LGTM - +1.

Not able to see a significant leap in perf on my quick tests - division doesn't seem to be a common scenario in my tests.

> Loop optimization for SIMD in ColumnDivideColumn.txt
> ----------------------------------------------------
>
>                 Key: HIVE-10235
>                 URL: https://issues.apache.org/jira/browse/HIVE-10235
>             Project: Hive
>          Issue Type: Sub-task
>          Components: Vectorization
>    Affects Versions: 1.1.0
>            Reporter: Chengxiang Li
>            Assignee: Chengxiang Li
>            Priority: Minor
>         Attachments: HIVE-10235.1.patch, HIVE-10235.1.patch
>
>
> Found two loop which could be optimized for packed instruction set during execution.
> 1. hasDivBy0 depends on the result of last loop, which prevent the loop be executed vectorized.
> {code:java}
> for(int i = 0; i != n; i++) {
>       <OperandType2> denom = vector2[i];
>       outputVector[i] = vector1[0] <OperatorSymbol> denom;
>       hasDivBy0 = hasDivBy0 || (denom == 0);
> }
> {code}
> 2. same as HIVE-10180, vector2\[0\] reference provent JVM optimizing loop into packed instruction set.
> {code:java}
> for(int i = 0; i != n; i++) {
>       outputVector[i] = vector1[i] <OperatorSymbol> vector2[0];
> }
> {code}



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