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 81E15C768 for ; Sat, 15 Nov 2014 17:07:34 +0000 (UTC) Received: (qmail 62680 invoked by uid 500); 15 Nov 2014 17:07:34 -0000 Delivered-To: apmail-hive-dev-archive@hive.apache.org Received: (qmail 62607 invoked by uid 500); 15 Nov 2014 17:07:33 -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 62596 invoked by uid 500); 15 Nov 2014 17:07:33 -0000 Delivered-To: apmail-hadoop-hive-dev@hadoop.apache.org Received: (qmail 62593 invoked by uid 99); 15 Nov 2014 17:07:33 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 15 Nov 2014 17:07:33 +0000 Date: Sat, 15 Nov 2014 17:07:33 +0000 (UTC) From: "Jimmy Xiang (JIRA)" To: hive-dev@hadoop.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Updated] (HIVE-8844) Choose a persisent policy for RDD caching [Spark Branch] 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-8844?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jimmy Xiang updated HIVE-8844: ------------------------------ Attachment: HIVE-8844.3-spark.patch Per Xuefu's suggestion, we'd better not to make it configurable for now since Hive has already too many configurations. So attached patch v3. > Choose a persisent policy for RDD caching [Spark Branch] > -------------------------------------------------------- > > Key: HIVE-8844 > URL: https://issues.apache.org/jira/browse/HIVE-8844 > Project: Hive > Issue Type: Sub-task > Components: Spark > Reporter: Xuefu Zhang > Assignee: Jimmy Xiang > Fix For: spark-branch > > Attachments: HIVE-8844.1-spark.patch, HIVE-8844.2-spark.patch, HIVE-8844.3-spark.patch > > > RDD caching is used for performance reasons in some multi-insert queries. Currently, we call RDD.cache(), which indicates a persistency policy of using memory only. We should choose a better policy. I think memory+disk will be good enough. Refer to RDD.persist() for more information. -- This message was sent by Atlassian JIRA (v6.3.4#6332)