[jira] [Work logged] (CRUNCH-679) Improvements for usage of DistCp
[
https://issues.apache.org/jira/browse/CRUNCH-679?focusedWorklogId=207918=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-207918
]
ASF GitHub Bot logged work on CRUNCH-679:
-
Author: ASF GitHub Bot
Created on: 05/Mar/19 18:03
Start Date: 05/Mar/19 18:03
Worklog Time Spent: 10m
Work Description: noslowerdna commented on pull request #20: CRUNCH-679:
Improvements for usage of DistCp
URL: https://github.com/apache/crunch/pull/20#discussion_r262615465
##
File path:
crunch-core/src/main/java/org/apache/crunch/util/CrunchRenameCopyListing.java
##
@@ -0,0 +1,269 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information regarding copyright
ownership. The ASF licenses this file to you 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.
+ */
+package org.apache.crunch.util;
+
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.FileStatus;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.io.SequenceFile;
+import org.apache.hadoop.io.Text;
+import org.apache.hadoop.security.Credentials;
+import org.apache.hadoop.tools.CopyListing;
+import org.apache.hadoop.tools.CopyListingFileStatus;
+import org.apache.hadoop.tools.DistCpOptions;
+import org.apache.hadoop.tools.DistCpOptions.FileAttribute;
+import org.apache.hadoop.tools.SimpleCopyListing;
+import org.apache.hadoop.tools.util.DistCpUtils;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Map;
+import java.util.Stack;
+
+/**
+ * A custom {@link CopyListing} implementation capable of dynamically renaming
+ * the target paths according to a configured set of values.
+ *
+ * Once https://issues.apache.org/jira/browse/HADOOP-16147 is available, this
+ * class can be significantly simplified.
+ *
+ */
+public class CrunchRenameCopyListing extends SimpleCopyListing {
+ /**
+ * Comma-separated list of original-file:renamed-file path rename pairs.
+ */
+ public static final String DISTCP_PATH_RENAMES =
"crunch.distcp.path.renames";
+
+ private static final Logger LOG =
LoggerFactory.getLogger(CrunchRenameCopyListing.class);
+ private final Map pathRenames;
+
+ private long totalPaths = 0;
+ private long totalBytesToCopy = 0;
+
+ /**
+ * Protected constructor, to initialize configuration.
+ *
+ * @param configuration The input configuration, with which the
source/target FileSystems may be accessed.
+ * @param credentials - Credentials object on which the FS delegation tokens
are cached. If null
+ * delegation token caching is skipped
+ */
+ protected CrunchRenameCopyListing(Configuration configuration, Credentials
credentials) {
+super(configuration, credentials);
+
+pathRenames = new HashMap<>();
+
+String[] pathRenameConf = configuration.getStrings(DISTCP_PATH_RENAMES);
Review comment:
I ran some tests and did not encounter any problems I would consider
showstopping. My largest concern is it appeared that the DistCp job completion
time was delayed roughly proportionally to the size of the configuration, with
a delay of nearly two minutes observed in the worst case test - I think
probably due to persisting the configuration into job history.
```
19/03/05 11:04:53 INFO mapreduce.Job: map 100% reduce 0%
19/03/05 11:06:48 INFO mapreduce.Job: Job job_1538164922410_717171 completed
successfully
```
Otherwise everything looked good, and viewing the job configuration through
the web UI worked fine although slow when the size was in multiple MBs.
The configuration property value size was retrieved by,
```
conf.get(CrunchRenameCopyListing.DISTCP_PATH_RENAMES).length()
```
Number of part files | Configuration size | Completion delay
-- | -- | --
10k | 260kb | 0:05
50k | 1.3mb | 0:29
250k | 6.8mb | 0:54
500k | 13.8mb | 1:55
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
[GitHub] [crunch] noslowerdna commented on a change in pull request #20: CRUNCH-679: Improvements for usage of DistCp
noslowerdna commented on a change in pull request #20: CRUNCH-679: Improvements
for usage of DistCp
URL: https://github.com/apache/crunch/pull/20#discussion_r262615465
##
File path:
crunch-core/src/main/java/org/apache/crunch/util/CrunchRenameCopyListing.java
##
@@ -0,0 +1,269 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information regarding copyright
ownership. The ASF licenses this file to you 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.
+ */
+package org.apache.crunch.util;
+
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.FileStatus;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Path;
+import org.apache.hadoop.io.SequenceFile;
+import org.apache.hadoop.io.Text;
+import org.apache.hadoop.security.Credentials;
+import org.apache.hadoop.tools.CopyListing;
+import org.apache.hadoop.tools.CopyListingFileStatus;
+import org.apache.hadoop.tools.DistCpOptions;
+import org.apache.hadoop.tools.DistCpOptions.FileAttribute;
+import org.apache.hadoop.tools.SimpleCopyListing;
+import org.apache.hadoop.tools.util.DistCpUtils;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.io.IOException;
+import java.util.HashMap;
+import java.util.Map;
+import java.util.Stack;
+
+/**
+ * A custom {@link CopyListing} implementation capable of dynamically renaming
+ * the target paths according to a configured set of values.
+ *
+ * Once https://issues.apache.org/jira/browse/HADOOP-16147 is available, this
+ * class can be significantly simplified.
+ *
+ */
+public class CrunchRenameCopyListing extends SimpleCopyListing {
+ /**
+ * Comma-separated list of original-file:renamed-file path rename pairs.
+ */
+ public static final String DISTCP_PATH_RENAMES =
"crunch.distcp.path.renames";
+
+ private static final Logger LOG =
LoggerFactory.getLogger(CrunchRenameCopyListing.class);
+ private final Map pathRenames;
+
+ private long totalPaths = 0;
+ private long totalBytesToCopy = 0;
+
+ /**
+ * Protected constructor, to initialize configuration.
+ *
+ * @param configuration The input configuration, with which the
source/target FileSystems may be accessed.
+ * @param credentials - Credentials object on which the FS delegation tokens
are cached. If null
+ * delegation token caching is skipped
+ */
+ protected CrunchRenameCopyListing(Configuration configuration, Credentials
credentials) {
+super(configuration, credentials);
+
+pathRenames = new HashMap<>();
+
+String[] pathRenameConf = configuration.getStrings(DISTCP_PATH_RENAMES);
Review comment:
I ran some tests and did not encounter any problems I would consider
showstopping. My largest concern is it appeared that the DistCp job completion
time was delayed roughly proportionally to the size of the configuration, with
a delay of nearly two minutes observed in the worst case test - I think
probably due to persisting the configuration into job history.
```
19/03/05 11:04:53 INFO mapreduce.Job: map 100% reduce 0%
19/03/05 11:06:48 INFO mapreduce.Job: Job job_1538164922410_717171 completed
successfully
```
Otherwise everything looked good, and viewing the job configuration through
the web UI worked fine although slow when the size was in multiple MBs.
The configuration property value size was retrieved by,
```
conf.get(CrunchRenameCopyListing.DISTCP_PATH_RENAMES).length()
```
Number of part files | Configuration size | Completion delay
-- | -- | --
10k | 260kb | 0:05
50k | 1.3mb | 0:29
250k | 6.8mb | 0:54
500k | 13.8mb | 1:55
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
