Homec4science
Diffusion elastic-yarn (master) Browse

R3704/hadoop-2.3.0-src/hadoop-tools/hadoop-distcp/src/site/xdoc/architecture.xml
2e9ad393f3f8master

Last Change
Raw File

architecture.xml
View Options

<?xml version="1.0" encoding="UTF-8"?>
<!--
Copyright 2002-2004 The Apache Software Foundation
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
<document xmlns="http://maven.apache.org/XDOC/2.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/XDOC/2.0 http://maven.apache.org/xsd/xdoc-2.0.xsd">
<head>
<title>Architecture of DistCp</title>
</head>
<body>
<section name="Architecture">
<p>The components of the new DistCp may be classified into the following
categories: </p>
<ul>
<li>DistCp Driver</li>
<li>Copy-listing generator</li>
<li>Input-formats and Map-Reduce components</li>
</ul>
<subsection name="DistCp Driver">
<p>The DistCp Driver components are responsible for:</p>
<ul>
<li>Parsing the arguments passed to the DistCp command on the
command-line, via:
<ul>
<li>OptionsParser, and</li>
<li>DistCpOptionsSwitch</li>
</ul>
</li>
<li>Assembling the command arguments into an appropriate
DistCpOptions object, and initializing DistCp. These arguments
include:
<ul>
<li>Source-paths</li>
<li>Target location</li>
<li>Copy options (e.g. whether to update-copy, overwrite, which
file-attributes to preserve, etc.)</li>
</ul>
</li>
<li>Orchestrating the copy operation by:
<ul>
<li>Invoking the copy-listing-generator to create the list of
files to be copied.</li>
<li>Setting up and launching the Hadoop Map-Reduce Job to carry
out the copy.</li>
<li>Based on the options, either returning a handle to the
Hadoop MR Job immediately, or waiting till completion.</li>
</ul>
</li>
</ul>
<br/>
<p>The parser-elements are exercised only from the command-line (or if
DistCp::run() is invoked). The DistCp class may also be used
programmatically, by constructing the DistCpOptions object, and
initializing a DistCp object appropriately.</p>
</subsection>
<subsection name="Copy-listing generator">
<p>The copy-listing-generator classes are responsible for creating the
list of files/directories to be copied from source. They examine
the contents of the source-paths (files/directories, including
wild-cards), and record all paths that need copy into a sequence-
file, for consumption by the DistCp Hadoop Job. The main classes in
this module include:</p>
<ol>
<li>CopyListing: The interface that should be implemented by any
copy-listing-generator implementation. Also provides the factory
method by which the concrete CopyListing implementation is
chosen.</li>
<li>SimpleCopyListing: An implementation of CopyListing that accepts
multiple source paths (files/directories), and recursively lists
all the individual files and directories under each, for
copy.</li>
<li>GlobbedCopyListing: Another implementation of CopyListing that
expands wild-cards in the source paths.</li>
<li>FileBasedCopyListing: An implementation of CopyListing that
reads the source-path list from a specified file.</li>
</ol>
<p/>
<p>Based on whether a source-file-list is specified in the
DistCpOptions, the source-listing is generated in one of the
following ways:</p>
<ol>
<li>If there's no source-file-list, the GlobbedCopyListing is used.
All wild-cards are expanded, and all the expansions are
forwarded to the SimpleCopyListing, which in turn constructs the
listing (via recursive descent of each path). </li>
<li>If a source-file-list is specified, the FileBasedCopyListing is
used. Source-paths are read from the specified file, and then
forwarded to the GlobbedCopyListing. The listing is then
constructed as described above.</li>
</ol>
<br/>
<p>One may customize the method by which the copy-listing is
constructed by providing a custom implementation of the CopyListing
interface. The behaviour of DistCp differs here from the legacy
DistCp, in how paths are considered for copy. </p>
<p>The legacy implementation only lists those paths that must
definitely be copied on to target.
E.g. if a file already exists at the target (and -overwrite isn't
specified), the file isn't even considered in the Map-Reduce Copy
Job. Determining this during setup (i.e. before the Map-Reduce Job)
involves file-size and checksum-comparisons that are potentially
time-consuming.</p>
<p>The new DistCp postpones such checks until the Map-Reduce Job, thus
reducing setup time. Performance is enhanced further since these
checks are parallelized across multiple maps.</p>
</subsection>
<subsection name="Input-formats and Map-Reduce components">
<p> The Input-formats and Map-Reduce components are responsible for
the actual copy of files and directories from the source to the
destination path. The listing-file created during copy-listing
generation is consumed at this point, when the copy is carried
out. The classes of interest here include:</p>
<ul>
<li><strong>UniformSizeInputFormat:</strong> This implementation of
org.apache.hadoop.mapreduce.InputFormat provides equivalence
with Legacy DistCp in balancing load across maps.
The aim of the UniformSizeInputFormat is to make each map copy
roughly the same number of bytes. Apropos, the listing file is
split into groups of paths, such that the sum of file-sizes in
each InputSplit is nearly equal to every other map. The splitting
isn't always perfect, but its trivial implementation keeps the
setup-time low.</li>
<li><strong>DynamicInputFormat and DynamicRecordReader:</strong>
<p> The DynamicInputFormat implements org.apache.hadoop.mapreduce.InputFormat,
and is new to DistCp. The listing-file is split into several
"chunk-files", the exact number of chunk-files being a multiple
of the number of maps requested for in the Hadoop Job. Each map
task is "assigned" one of the chunk-files (by renaming the chunk
to the task's id), before the Job is launched.</p>
<p>Paths are read from each chunk using the DynamicRecordReader,
and processed in the CopyMapper. After all the paths in a chunk
are processed, the current chunk is deleted and a new chunk is
acquired. The process continues until no more chunks are
available.</p>
<p>This "dynamic" approach allows faster map-tasks to consume
more paths than slower ones, thus speeding up the DistCp job
overall. </p>
</li>
<li><strong>CopyMapper:</strong> This class implements the physical
file-copy. The input-paths are checked against the input-options
(specified in the Job's Configuration), to determine whether a
file needs copy. A file will be copied only if at least one of
the following is true:
<ul>
<li>A file with the same name doesn't exist at target.</li>
<li>A file with the same name exists at target, but has a
different file size.</li>
<li>A file with the same name exists at target, but has a
different checksum, and -skipcrccheck isn't mentioned.</li>
<li>A file with the same name exists at target, but -overwrite
is specified.</li>
<li>A file with the same name exists at target, but differs in
block-size (and block-size needs to be preserved.</li>
</ul>
</li>
<li><strong>CopyCommitter:</strong>
This class is responsible for the commit-phase of the DistCp
job, including:
<ul>
<li>Preservation of directory-permissions (if specified in the
options)</li>
<li>Clean-up of temporary-files, work-directories, etc.</li>
</ul>
</li>
</ul>
</subsection>
</section>
</body>
</document>
c4science · Help