From mapreduce-issues-return-24426-apmail-hadoop-mapreduce-issues-archive=hadoop.apache.org@hadoop.apache.org Fri Aug 26 19:35:55 2011 Return-Path: X-Original-To: apmail-hadoop-mapreduce-issues-archive@minotaur.apache.org Delivered-To: apmail-hadoop-mapreduce-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 712308EA1 for ; Fri, 26 Aug 2011 19:35:55 +0000 (UTC) Received: (qmail 81759 invoked by uid 500); 26 Aug 2011 19:35:55 -0000 Delivered-To: apmail-hadoop-mapreduce-issues-archive@hadoop.apache.org Received: (qmail 81719 invoked by uid 500); 26 Aug 2011 19:35:54 -0000 Mailing-List: contact mapreduce-issues-help@hadoop.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: mapreduce-issues@hadoop.apache.org Delivered-To: mailing list mapreduce-issues@hadoop.apache.org Received: (qmail 81711 invoked by uid 99); 26 Aug 2011 19:35:54 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 26 Aug 2011 19:35:54 +0000 X-ASF-Spam-Status: No, hits=-2000.9 required=5.0 tests=ALL_TRUSTED,RP_MATCHES_RCVD X-Spam-Check-By: apache.org Received: from [140.211.11.116] (HELO hel.zones.apache.org) (140.211.11.116) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 26 Aug 2011 19:35:51 +0000 Received: from hel.zones.apache.org (hel.zones.apache.org [140.211.11.116]) by hel.zones.apache.org (Postfix) with ESMTP id BE90FD1EF4 for ; Fri, 26 Aug 2011 19:35:29 +0000 (UTC) Date: Fri, 26 Aug 2011 19:35:29 +0000 (UTC) From: "Alejandro Abdelnur (JIRA)" To: mapreduce-issues@hadoop.apache.org Message-ID: <1999508288.18900.1314387329777.JavaMail.tomcat@hel.zones.apache.org> In-Reply-To: <1489319984.1211.1312286727371.JavaMail.tomcat@hel.zones.apache.org> Subject: [jira] [Commented] (MAPREDUCE-2765) DistCp Rewrite MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 X-Virus-Checked: Checked by ClamAV on apache.org [ https://issues.apache.org/jira/browse/MAPREDUCE-2765?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13091976#comment-13091976 ] Alejandro Abdelnur commented on MAPREDUCE-2765: ----------------------------------------------- Comments on the POM: * 1 It should use hadoop-project POM as parent * 2 Version should be hardcoded, not property * 3 It should not define repositories (they are defined in hadoop-project POM) * 4 All dependencies should not have versions, thus taking the version from the hadoop-project POM dependenciesManagement section (if there is a new dependency introduced by distcp, it should be added to the hadoop-project POM dependenciesManagement section and in this POM without version as the rest, this ensure across all Hadoop modules the same versions of all JARs are used at compile/test/dist time) * 5 Why aspect JARs have compile scope (no scope means compile scope)? * 6 Similarly Plugin versions should be defined in the pluginManagement section in the hadoop-project POM * 7 For checkstyle/findbugs/clover see to reuse the settings of hadoop-project POM * 8 clover license is already defined in hadoop-project POM (and the one here seems an internal Y! location) * 9 The surefire plugin section, you should use the syntax PROP_VALUE (see hadoop-project POM) * 10 It seems the dependency plugin is being used to create the tool dist in the target/ dir. This is not correct, an assembly should be used. Ideally the existing hadoop-dist assembly should be reused. #10 brings up to the table how do we want to run tools? How their classpath/librarypath is created, which JARs are used bundled with the tool, etc. Finally, based on the outcome of the discussion of where tools should be (trunk/hadoop-tools-project/ or at /trunk/hadoop-*-project/hadoop-*-tools) the patch should be updated to it. > DistCp Rewrite > -------------- > > Key: MAPREDUCE-2765 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-2765 > Project: Hadoop Map/Reduce > Issue Type: New Feature > Components: distcp > Affects Versions: 0.20.203.0 > Reporter: Mithun Radhakrishnan > Assignee: Mithun Radhakrishnan > Attachments: distcpv2.20.203.patch, distcpv2_trunk.patch, distcpv2_trunk_post_review_1.patch > > > This is a slightly modified version of the DistCp rewrite that Yahoo uses in production today. The rewrite was ground-up, with specific focus on: > 1. improved startup time (postponing as much work as possible to the MR job) > 2. support for multiple copy-strategies > 3. new features (e.g. -atomic, -async, -bandwidth.) > 4. improved programmatic use > Some effort has gone into refactoring what used to be achieved by a single large (1.7 KLOC) source file, into a design that (hopefully) reads better too. > The proposed DistCpV2 preserves command-line-compatibility with the old version, and should be a drop-in replacement. > New to v2: > 1. Copy-strategies and the DynamicInputFormat: > A copy-strategy determines the policy by which source-file-paths are distributed between map-tasks. (These boil down to the choice of the input-format.) > If no strategy is explicitly specified on the command-line, the policy chosen is "uniform size", where v2 behaves identically to old-DistCp. (The number of bytes transferred by each map-task is roughly equal, at a per-file granularity.) > Alternatively, v2 ships with a "dynamic" copy-strategy (in the DynamicInputFormat). This policy acknowledges that > (a) dividing files based only on file-size might not be an even distribution (E.g. if some datanodes are slower than others, or if some files are skipped.) > (b) a "static" association of a source-path to a map increases the likelihood of long-tails during copy. > The "dynamic" strategy divides the list-of-source-paths into a number (> nMaps) of smaller parts. When each map completes its current list of paths, it picks up a new list to process, if available. So if a map-task is stuck on a slow (and not necessarily large) file, other maps can pick up the slack. The thinner the file-list is sliced, the greater the parallelism (and the lower the chances of long-tails). Within reason, of course: the number of these short-lived list-files is capped at an overridable maximum. > Internal benchmarks against source/target clusters with some slow(ish) datanodes have indicated significant performance gains when using the dynamic-strategy. Gains are most pronounced when nFiles greatly exceeds nMaps. > Please note that the DynamicInputFormat might prove useful outside of DistCp. It is hence available as a mapred/lib, unfettered to DistCpV2. Also note that the copy-strategies have no bearing on the CopyMapper.map() implementation. > > 2. Improved startup-time and programmatic use: > When the old-DistCp runs with -update, and creates the list-of-source-paths, it attempts to filter out files that might be skipped (by comparing file-sizes, checksums, etc.) This significantly increases the startup time (or the time spent in serial processing till the MR job is launched), blocking the calling-thread. This becomes pronounced as nFiles increases. (Internal benchmarks have seen situations where more time is spent setting up the job than on the actual transfer.) > DistCpV2 postpones as much work as possible to the MR job. The file-listing isn't filtered until the map-task runs (at which time, identical files are skipped). DistCpV2 can now be run "asynchronously". The program quits at job-launch, logging the job-id for tracking. Programmatically, the DistCp.execute() returns a Job instance for progress-tracking. > > 3. New features: > (a) -async: As described in #2. > (b) -atomic: Data is copied to a (user-specifiable) tmp-location, and then moved atomically to destination. > (c) -bandwidth: Enforces a limit on the bandwidth consumed per map. > (d) -strategy: As above. > > A more comprehensive description the newer features, how the dynamic-strategy works, etc. is available in src/site/xdoc/, and in the pdf that's generated therefrom, during the build. > High on the list of things to do is support to parallelize copies on a per-block level. (i.e. Incorporation of HDFS-222.) > I look forward to comments, suggestions and discussion that will hopefully ensue. I have this running against Hadoop 0.20.203.0. I also have a port to 0.23.0 (complete with unit-tests). > P.S. > A tip of the hat to Srikanth (Sundarrajan) and Venkatesh (Seetharamaiah), for ideas, code, reviews and guidance. Although much of the code is mine, the idea to use the DFS to implement "dynamic" input-splits wasn't. > -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira