HBASE-13303 Fix size calculation of results on the region server

This commit is contained in:
Andrew Purtell 2015-03-20 17:24:05 -07:00
parent 0915191167
commit 8dfed5dc88
4 changed files with 165 additions and 2 deletions

View File

@ -568,6 +568,21 @@ public final class CellUtil {
return estimatedSerializedSizeOf(cell);
}
/**
* This is a hack that should be removed once we don't care about matching
* up client- and server-side estimations of cell size. It needed to be
* backwards compatible with estimations done by older clients. We need to
* pretend that tags never exist and cells aren't serialized with tag
* length included. See HBASE-13262 and HBASE-13303
*/
@Deprecated
public static long estimatedHeapSizeOfWithoutTags(final Cell cell) {
if (cell instanceof KeyValue) {
return ((KeyValue)cell).heapSizeWithoutTags();
}
return getSumOfCellKeyElementLengths(cell) + cell.getValueLength();
}
/********************* tags *************************************/
/**
* Util method to iterate through the tags

View File

@ -2640,6 +2640,27 @@ public class KeyValue implements Cell, HeapSize, Cloneable, SettableSequenceId,
return ClassSize.align(sum);
}
/**
* This is a hack that should be removed once we don't care about matching
* up client- and server-side estimations of cell size. It needed to be
* backwards compatible with estimations done by older clients. We need to
* pretend that tags never exist and KeyValues aren't serialized with tag
* length included. See HBASE-13262 and HBASE-13303
*/
@Deprecated
public long heapSizeWithoutTags() {
int sum = 0;
sum += ClassSize.OBJECT;// the KeyValue object itself
sum += ClassSize.REFERENCE;// pointer to "bytes"
sum += ClassSize.align(ClassSize.ARRAY);// "bytes"
sum += KeyValue.KEYVALUE_INFRASTRUCTURE_SIZE;
sum += getKeyLength();
sum += getValueLength();
sum += 2 * Bytes.SIZEOF_INT;// offset, length
sum += Bytes.SIZEOF_LONG;// memstoreTS
return ClassSize.align(sum);
}
/**
* A simple form of KeyValue that creates a keyvalue with only the key part of the byte[]
* Mainly used in places where we need to compare two cells. Avoids copying of bytes

View File

@ -2129,7 +2129,7 @@ public class RSRpcServices implements HBaseRPCErrorHandler,
for (Result r : results) {
for (Cell cell : r.rawCells()) {
totalCellSize += CellUtil.estimatedSerializedSizeOf(cell);
currentScanResultSize += CellUtil.estimatedHeapSizeOf(cell);
currentScanResultSize += CellUtil.estimatedHeapSizeOfWithoutTags(cell);
}
}
}
@ -2192,7 +2192,7 @@ public class RSRpcServices implements HBaseRPCErrorHandler,
// If the calculation can't be skipped, then do it now.
if (!skipResultSizeCalculation) {
currentScanResultSize += CellUtil.estimatedHeapSizeOf(cell);
currentScanResultSize += CellUtil.estimatedHeapSizeOfWithoutTags(cell);
}
}
// The size limit was reached. This means there are more cells remaining in

View File

@ -0,0 +1,127 @@
/*
* 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.hadoop.hbase.client;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.hbase.HBaseTestingUtility;
import org.apache.hadoop.hbase.HConstants;
import org.apache.hadoop.hbase.KeyValue;
import org.apache.hadoop.hbase.TableName;
import org.apache.hadoop.hbase.Tag;
import org.apache.hadoop.hbase.client.HTable;
import org.apache.hadoop.hbase.client.Put;
import org.apache.hadoop.hbase.client.ResultScanner;
import org.apache.hadoop.hbase.client.Scan;
import org.apache.hadoop.hbase.io.hfile.HFile;
import org.apache.hadoop.hbase.testclassification.LargeTests;
import org.apache.hadoop.hbase.util.Bytes;
import org.junit.AfterClass;
import org.junit.BeforeClass;
import org.junit.Test;
import org.junit.experimental.categories.Category;
import static org.junit.Assert.assertEquals;
@Category(LargeTests.class)
public class TestResultSizeEstimation {
final static HBaseTestingUtility TEST_UTIL = new HBaseTestingUtility();
final static int TAG_DATA_SIZE = 2048;
final static int SCANNER_DATA_LIMIT = TAG_DATA_SIZE + 256;
@BeforeClass
public static void setUpBeforeClass() throws Exception {
Configuration conf = TEST_UTIL.getConfiguration();
// Need HFileV3
conf.setInt(HFile.FORMAT_VERSION_KEY, HFile.MIN_FORMAT_VERSION_WITH_TAGS);
// effectively limit max result size to one entry if it has tags
conf.setLong(HConstants.HBASE_CLIENT_SCANNER_MAX_RESULT_SIZE_KEY, SCANNER_DATA_LIMIT);
conf.setBoolean(ScannerCallable.LOG_SCANNER_ACTIVITY, true);
TEST_UTIL.startMiniCluster(1);
}
@AfterClass
public static void tearDownAfterClass() throws Exception {
TEST_UTIL.shutdownMiniCluster();
}
@Test
public void testResultSizeEstimation() throws Exception {
byte [] ROW1 = Bytes.toBytes("testRow1");
byte [] ROW2 = Bytes.toBytes("testRow2");
byte [] FAMILY = Bytes.toBytes("testFamily");
byte [] QUALIFIER = Bytes.toBytes("testQualifier");
byte [] VALUE = Bytes.toBytes("testValue");
TableName TABLE = TableName.valueOf("testResultSizeEstimation");
byte[][] FAMILIES = new byte[][] { FAMILY };
Configuration conf = new Configuration(TEST_UTIL.getConfiguration());
HTable table = TEST_UTIL.createTable(TABLE, FAMILIES, conf);
Put p = new Put(ROW1);
p.add(new KeyValue(ROW1, FAMILY, QUALIFIER, Long.MAX_VALUE, VALUE));
table.put(p);
p = new Put(ROW2);
p.add(new KeyValue(ROW2, FAMILY, QUALIFIER, Long.MAX_VALUE, VALUE));
table.put(p);
Scan s = new Scan();
s.setMaxResultSize(SCANNER_DATA_LIMIT);
ResultScanner rs = table.getScanner(s);
int count = 0;
while(rs.next() != null) {
count++;
}
assertEquals("Result size estimation did not work properly", 2, count);
rs.close();
table.close();
}
@Test
public void testResultSizeEstimationWithTags() throws Exception {
byte [] ROW1 = Bytes.toBytes("testRow1");
byte [] ROW2 = Bytes.toBytes("testRow2");
byte [] FAMILY = Bytes.toBytes("testFamily");
byte [] QUALIFIER = Bytes.toBytes("testQualifier");
byte [] VALUE = Bytes.toBytes("testValue");
TableName TABLE = TableName.valueOf("testResultSizeEstimationWithTags");
byte[][] FAMILIES = new byte[][] { FAMILY };
Configuration conf = new Configuration(TEST_UTIL.getConfiguration());
HTable table = TEST_UTIL.createTable(TABLE, FAMILIES, conf);
Put p = new Put(ROW1);
p.add(new KeyValue(ROW1, FAMILY, QUALIFIER, Long.MAX_VALUE, VALUE,
new Tag[] { new Tag((byte)1, new byte[TAG_DATA_SIZE]) } ));
table.put(p);
p = new Put(ROW2);
p.add(new KeyValue(ROW2, FAMILY, QUALIFIER, Long.MAX_VALUE, VALUE,
new Tag[] { new Tag((byte)1, new byte[TAG_DATA_SIZE]) } ));
table.put(p);
Scan s = new Scan();
s.setMaxResultSize(SCANNER_DATA_LIMIT);
ResultScanner rs = table.getScanner(s);
int count = 0;
while(rs.next() != null) {
count++;
}
assertEquals("Result size estimation did not work properly", 2, count);
rs.close();
table.close();
}
}