Return-Path: X-Original-To: apmail-hive-dev-archive@www.apache.org Delivered-To: apmail-hive-dev-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 9209310872 for ; Fri, 23 Aug 2013 23:45:25 +0000 (UTC) Received: (qmail 73661 invoked by uid 500); 23 Aug 2013 23:45:25 -0000 Delivered-To: apmail-hive-dev-archive@hive.apache.org Received: (qmail 73616 invoked by uid 500); 23 Aug 2013 23:45:25 -0000 Mailing-List: contact dev-help@hive.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@hive.apache.org Delivered-To: mailing list dev@hive.apache.org Received: (qmail 73605 invoked by uid 99); 23 Aug 2013 23:45:25 -0000 Received: from reviews-vm.apache.org (HELO reviews.apache.org) (140.211.11.40) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 23 Aug 2013 23:45:25 +0000 Received: from reviews.apache.org (localhost [127.0.0.1]) by reviews.apache.org (Postfix) with ESMTP id E1BBC1D2E0F; Fri, 23 Aug 2013 23:45:23 +0000 (UTC) Content-Type: multipart/alternative; boundary="===============4383146470889312424==" MIME-Version: 1.0 Subject: Review Request 13787: HIVE-5095: Hive needs new operator walker for parallelization/optimization for tez From: "Vikram Dixit Kumaraswamy" To: "Gunther Hagleitner" Cc: "Vikram Dixit Kumaraswamy" , "hive" Date: Fri, 23 Aug 2013 23:45:23 -0000 Message-ID: <20130823234523.19978.98150@reviews.apache.org> X-ReviewBoard-URL: https://reviews.apache.org Auto-Submitted: auto-generated Sender: "Vikram Dixit Kumaraswamy" X-ReviewGroup: hive X-ReviewRequest-URL: https://reviews.apache.org/r/13787/ X-Sender: "Vikram Dixit Kumaraswamy" Reply-To: "Vikram Dixit Kumaraswamy" --===============4383146470889312424== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13787/ ----------------------------------------------------------- Review request for hive and Gunther Hagleitner. Bugs: HIVE-5095 https://issues.apache.org/jira/browse/HIVE-5095 Repository: hive-git Description ------- For tez to compute the number of reducers, we should be walking the operator tree in a topological fashion so that the reducers down the tree get the estimate from all parents. However, the current walkers in hive only walk the operator tree in a depth-first fashion. We need to add a new walker for the topological walk. Also, since information about the parent operators needs to be propagated on a per parent basis, we need to retain some context across operators to be passed to the child which the walker will co-ordinate. Diffs ----- common/src/java/org/apache/hadoop/hive/conf/HiveConf.java 7408a5a ql/src/java/org/apache/hadoop/hive/ql/exec/Operator.java ca48f5e ql/src/java/org/apache/hadoop/hive/ql/exec/ReduceSinkOperator.java 6a538e8 ql/src/java/org/apache/hadoop/hive/ql/exec/TableScanOperator.java 6ee13ec ql/src/java/org/apache/hadoop/hive/ql/exec/Utilities.java f3c34d1 ql/src/java/org/apache/hadoop/hive/ql/exec/mr/MapRedTask.java 49a0ee3 ql/src/java/org/apache/hadoop/hive/ql/optimizer/GenMapRedUtils.java 7433ddc ql/src/java/org/apache/hadoop/hive/ql/optimizer/OpProcessor.java PRE-CREATION ql/src/java/org/apache/hadoop/hive/ql/optimizer/physical/OpProcContext.java PRE-CREATION ql/src/java/org/apache/hadoop/hive/ql/parse/GenOpGraphWalker.java PRE-CREATION ql/src/java/org/apache/hadoop/hive/ql/parse/GenTezProcContext.java 827637a ql/src/java/org/apache/hadoop/hive/ql/parse/GenTezWork.java ff8b17b ql/src/java/org/apache/hadoop/hive/ql/parse/MapReduceCompiler.java c1c1da5 ql/src/java/org/apache/hadoop/hive/ql/parse/TaskCompiler.java 248eb03 ql/src/java/org/apache/hadoop/hive/ql/parse/TezCompiler.java 5abedfe ql/src/java/org/apache/hadoop/hive/ql/plan/PlanUtils.java 5fd8d828 Diff: https://reviews.apache.org/r/13787/diff/ Testing ------- Thanks, Vikram Dixit Kumaraswamy --===============4383146470889312424==--