-
Notifications
You must be signed in to change notification settings - Fork 483
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
ORC-1017: Add sizes tool to determine and display the sizes of each c…
…olumn in a set of files. (#925) ### What changes were proposed in this pull request? This patch adds a new tool that accounts for the total size of a set of ORC files. For files written by >= ORC 1.5, you'll get a column breakdown of the file. There are some virtual columns that are included: - _index the indexes that are used for skipping inside the stripe - _data the data in files written prior to ORC 1.5 - _stripe_footer the stripe metadata - _file_footer the file metadata - _padding padding added to align stripes to HDFS block boundaries I also added a new method on TypeDescription that gets the full field name, which is the inverse of findSubtype. ### Why are the changes needed? The tool helps diagnose the compression of a set of files. ### How was this patch tested? I added a test of the new TypeDescription.getFullFieldName. I ran the tool over some of the examples and some multiple-terabyte directories of production ORC files. (cherry picked from commit be0762b) Signed-off-by: Dongjoon Hyun <[email protected]>
- Loading branch information
1 parent
43b53ea
commit 0fbfa66
Showing
6 changed files
with
336 additions
and
26 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
196 changes: 196 additions & 0 deletions
196
java/tools/src/java/org/apache/orc/tools/ColumnSizes.java
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,196 @@ | ||
| /* | ||
| * 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.orc.tools; | ||
|
|
||
| import org.apache.hadoop.conf.Configuration; | ||
| import org.apache.hadoop.fs.FileSystem; | ||
| import org.apache.hadoop.fs.LocatedFileStatus; | ||
| import org.apache.hadoop.fs.Path; | ||
| import org.apache.hadoop.fs.RemoteIterator; | ||
| import org.apache.orc.ColumnStatistics; | ||
| import org.apache.orc.OrcFile; | ||
| import org.apache.orc.Reader; | ||
| import org.apache.orc.StripeInformation; | ||
| import org.apache.orc.TypeDescription; | ||
|
|
||
| import java.io.IOException; | ||
| import java.io.PrintStream; | ||
| import java.util.ArrayList; | ||
| import java.util.List; | ||
|
|
||
| /** | ||
| * Given a set of paths, finds all of the "*.orc" files under them and | ||
| * prints the sizes of each column, both as a percentage and the number of | ||
| * bytes per a row. | ||
| */ | ||
| public class ColumnSizes { | ||
| final Configuration conf; | ||
| final TypeDescription schema; | ||
| final long[] columnSizes; | ||
| int goodFiles = 0; | ||
| long rows = 0; | ||
| long padding = 0; | ||
| long totalSize = 0; | ||
| long stripeFooterSize = 0; | ||
| long fileFooterSize = 0; | ||
| long stripeIndex = 0; | ||
| // data bytes that aren't assigned to a specific column | ||
| long stripeData = 0; | ||
|
|
||
| public ColumnSizes(Configuration conf, | ||
| LocatedFileStatus file) throws IOException { | ||
| this.conf = conf; | ||
| try (Reader reader = OrcFile.createReader(file.getPath(), | ||
| OrcFile.readerOptions(conf))) { | ||
| this.schema = reader.getSchema(); | ||
| columnSizes = new long[schema.getMaximumId() + 1]; | ||
| addReader(file, reader); | ||
| } | ||
| } | ||
|
|
||
| private void checkStripes(LocatedFileStatus file, | ||
| Reader reader) { | ||
| // Count the magic as file overhead | ||
| long offset = OrcFile.MAGIC.length(); | ||
| fileFooterSize += offset; | ||
|
|
||
| for (StripeInformation stripe: reader.getStripes()) { | ||
| padding += stripe.getOffset() - offset; | ||
| stripeIndex += stripe.getIndexLength(); | ||
| stripeData += stripe.getDataLength(); | ||
| stripeFooterSize += stripe.getFooterLength(); | ||
| offset = stripe.getOffset() + stripe.getLength(); | ||
| } | ||
| // Add everything else as the file footer | ||
| fileFooterSize += file.getLen() - offset; | ||
| } | ||
|
|
||
| private boolean addReader(LocatedFileStatus file, | ||
| Reader reader) { | ||
| // Validate that the schemas are the same | ||
| TypeDescription newSchema = reader.getSchema(); | ||
| if (schema.equals(newSchema)) { | ||
| goodFiles += 1; | ||
| rows += reader.getNumberOfRows(); | ||
| totalSize += file.getLen(); | ||
| checkStripes(file, reader); | ||
| ColumnStatistics[] colStats = reader.getStatistics(); | ||
| for (int c = 0; c < colStats.length && c < columnSizes.length; c++) { | ||
| columnSizes[c] += colStats[c].getBytesOnDisk(); | ||
| // Don't double count. Either count the bytes as stripe data or as | ||
| // part of a column. | ||
| stripeData -= colStats[c].getBytesOnDisk(); | ||
| } | ||
| } else { | ||
| System.err.println("Ignoring " + file.getPath() | ||
| + " because of schema mismatch: " + newSchema); | ||
| return false; | ||
| } | ||
| return true; | ||
| } | ||
|
|
||
| public boolean addFile(LocatedFileStatus file) throws IOException { | ||
| try (Reader reader = OrcFile.createReader(file.getPath(), | ||
| OrcFile.readerOptions(conf))) { | ||
| return addReader(file, reader); | ||
| } | ||
| } | ||
|
|
||
| private static class StringLongPair { | ||
| final String name; | ||
| final long size; | ||
| StringLongPair(String name, long size) { | ||
| this.name = name; | ||
| this.size = size; | ||
| } | ||
| } | ||
|
|
||
| private void printResults(PrintStream out) { | ||
| List<StringLongPair> sizes = new ArrayList<>(columnSizes.length + 5); | ||
| for(int column = 0; column < columnSizes.length; ++column) { | ||
| if (columnSizes[column] > 0) { | ||
| sizes.add(new StringLongPair( | ||
| schema.findSubtype(column).getFullFieldName(), | ||
| columnSizes[column])); | ||
| } | ||
| } | ||
| if (padding > 0) { | ||
| sizes.add(new StringLongPair("_padding", padding)); | ||
| } | ||
| if (stripeFooterSize > 0) { | ||
| sizes.add(new StringLongPair("_stripe_footer", stripeFooterSize)); | ||
| } | ||
| if (fileFooterSize > 0) { | ||
| sizes.add(new StringLongPair("_file_footer", fileFooterSize)); | ||
| } | ||
| if (stripeIndex > 0) { | ||
| sizes.add(new StringLongPair("_index", stripeIndex)); | ||
| } | ||
| if (stripeData > 0) { | ||
| sizes.add(new StringLongPair("_data", stripeData)); | ||
| } | ||
| // sort by descending size, ascending name | ||
| sizes.sort((x, y) -> x.size != y.size ? | ||
| Long.compare(y.size, x.size) : x.name.compareTo(y.name)); | ||
| out.println("Percent Bytes/Row Name"); | ||
| for (StringLongPair item: sizes) { | ||
| out.println(String.format(" %-5.2f %-9.2f %s", | ||
| 100.0 * item.size / totalSize, (double) item.size / rows, item.name)); | ||
| } | ||
| } | ||
|
|
||
| public static void main(Configuration conf, String[] args) throws IOException { | ||
| ColumnSizes result = null; | ||
| int badFiles = 0; | ||
| for(String root: args) { | ||
| Path rootPath = new Path(root); | ||
| FileSystem fs = rootPath.getFileSystem(conf); | ||
| for(RemoteIterator<LocatedFileStatus> itr = fs.listFiles(rootPath, true); itr.hasNext(); ) { | ||
| LocatedFileStatus status = itr.next(); | ||
| if (status.isFile() && status.getPath().getName().endsWith(".orc")) { | ||
| try { | ||
| if (result == null) { | ||
| result = new ColumnSizes(conf, status); | ||
| } else { | ||
| if (!result.addFile(status)) { | ||
| badFiles += 1; | ||
| } | ||
| } | ||
| } catch (IOException err) { | ||
| badFiles += 1; | ||
| System.err.println("Failed to read " + status.getPath()); | ||
| } | ||
| } | ||
| } | ||
| } | ||
| if (result == null) { | ||
| System.err.println("No files found"); | ||
| } else { | ||
| result.printResults(System.out); | ||
| } | ||
| if (badFiles > 0) { | ||
| System.err.println(badFiles + " bad ORC files found."); | ||
| System.exit(1); | ||
| } | ||
| } | ||
|
|
||
| public static void main(String[] args) throws IOException { | ||
| main(new Configuration(), args); | ||
| } | ||
| } |
Oops, something went wrong.