Apache
/
hadoop
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

HADOOP-13680. fs.s3a.readahead.range to use getLongBytes. Contributed by Abhishek Modi.

This commit is contained in:
Steve Loughran 2016-10-31 20:52:49 +00:00
parent f3eb4c3c73
commit a1761a841e
5 changed files with 73 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
hadoop-common-project/hadoop-common/src/main/resources/core-default.xml
View File

@ -1055,8 +1055,10 @@
<property>
<name>fs.s3a.multipart.size</name>
<value>104857600</value>
<description>How big (in bytes) to split upload or copy operations up into.</description>
<value>100M</value>
<description>How big (in bytes) to split upload or copy operations up into.
A suffix from the set {K,M,G,T,P} may be used to scale the numeric value.
</description>
</property>
<property>
@ -1064,7 +1066,8 @@
<value>2147483647</value>
<description>How big (in bytes) to split upload or copy operations up into.
This also controls the partition size in renamed files, as rename() involves
copying the source file(s)
copying the source file(s).
A suffix from the set {K,M,G,T,P} may be used to scale the numeric value.
</description>
</property>
@ -1120,8 +1123,9 @@
<property>
<name>fs.s3a.block.size</name>
<value>33554432</value>
<value>32M</value>
<description>Block size to use when reading files using s3a: file system.
A suffix from the set {K,M,G,T,P} may be used to scale the numeric value.
</description>
</property>
@ -1183,10 +1187,12 @@
<property>
<name>fs.s3a.readahead.range</name>
<value>65536</value>
<value>64K</value>
<description>Bytes to read ahead during a seek() before closing and
re-opening the S3 HTTP connection. This option will be overridden if
any call to setReadahead() is made to an open stream.</description>
any call to setReadahead() is made to an open stream.
A suffix from the set {K,M,G,T,P} may be used to scale the numeric value.
</description>
</property>
<property>

17
hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AFileSystem.java
View File

@ -183,10 +183,11 @@ public class S3AFileSystem extends FileSystem {
MIN_MULTIPART_THRESHOLD, DEFAULT_MIN_MULTIPART_THRESHOLD);
//check but do not store the block size
longOption(conf, FS_S3A_BLOCK_SIZE, DEFAULT_BLOCKSIZE, 1);
longBytesOption(conf, FS_S3A_BLOCK_SIZE, DEFAULT_BLOCKSIZE, 1);
enableMultiObjectsDelete = conf.getBoolean(ENABLE_MULTI_DELETE, true);
readAhead = longOption(conf, READAHEAD_RANGE, DEFAULT_READAHEAD_RANGE, 0);
readAhead = longBytesOption(conf, READAHEAD_RANGE,
DEFAULT_READAHEAD_RANGE, 0);
storageStatistics = (S3AStorageStatistics)
GlobalStorageStatistics.INSTANCE
.put(S3AStorageStatistics.NAME,
@ -357,6 +358,16 @@ public class S3AFileSystem extends FileSystem {
return s3;
}
/**
* Returns the read ahead range value used by this filesystem
* @return
*/
@VisibleForTesting
long getReadAheadRange() {
return readAhead;
}
/**
* Get the input policy for this FS instance.
* @return the input policy
@ -1883,7 +1894,7 @@ public class S3AFileSystem extends FileSystem {
*/
@Deprecated
public long getDefaultBlockSize() {
return getConf().getLong(FS_S3A_BLOCK_SIZE, DEFAULT_BLOCKSIZE);
return getConf().getLongBytes(FS_S3A_BLOCK_SIZE, DEFAULT_BLOCKSIZE);
}
@Override

23
hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AUtils.java
View File

@ -509,6 +509,27 @@ public final class S3AUtils {
return v;
}
/**
* Get a long option >= the minimum allowed value, supporting memory
* prefixes K,M,G,T,P.
* @param conf configuration
* @param key key to look up
* @param defVal default value
* @param min minimum value
* @return the value
* @throws IllegalArgumentException if the value is below the minimum
*/
static long longBytesOption(Configuration conf,
String key,
long defVal,
long min) {
long v = conf.getLongBytes(key, defVal);
Preconditions.checkArgument(v >= min,
String.format("Value of %s: %d is below the minimum value %d",
key, v, min));
return v;
}
/**
* Get a size property from the configuration: this property must
* be at least equal to {@link Constants#MULTIPART_MIN_SIZE}.
@ -521,7 +542,7 @@ public final class S3AUtils {
*/
public static long getMultipartSizeProperty(Configuration conf,
String property, long defVal) {
long partSize = conf.getLong(property, defVal);
long partSize = conf.getLongBytes(property, defVal);
if (partSize < MULTIPART_MIN_SIZE) {
LOG.warn("{} must be at least 5 MB; configured value is {}",
property, partSize);

22
hadoop-tools/hadoop-aws/src/site/markdown/tools/hadoop-aws/index.md
View File

@ -762,16 +762,20 @@ from placing its declaration on the command line.
<property>
<name>fs.s3a.multipart.size</name>
<value>104857600</value>
<value>100M</value>
<description>How big (in bytes) to split upload or copy operations up into.
This also controls the partition size in renamed files, as rename() involves
copying the source file(s)</description>
A suffix from the set {K,M,G,T,P} may be used to scale the numeric value.
</description>
</property>
<property>
<name>fs.s3a.multipart.threshold</name>
<value>2147483647</value>
<description>Threshold before uploads or copies use parallel multipart operations.</description>
<description>How big (in bytes) to split upload or copy operations up into.
This also controls the partition size in renamed files, as rename() involves
copying the source file(s).
A suffix from the set {K,M,G,T,P} may be used to scale the numeric value.
</description>
</property>
<property>
@ -825,7 +829,7 @@ from placing its declaration on the command line.
<property>
<name>fs.s3a.block.size</name>
<value>33554432</value>
<value>32M</value>
<description>Block size to use when reading files using s3a: file system.
</description>
</property>
@ -859,7 +863,7 @@ from placing its declaration on the command line.
<property>
<name>fs.s3a.readahead.range</name>
<value>65536</value>
<value>64K</value>
<description>Bytes to read ahead during a seek() before closing and
re-opening the S3 HTTP connection. This option will be overridden if
any call to setReadahead() is made to an open stream.</description>
@ -1029,9 +1033,9 @@ S3 endpoints, as disks are not used for intermediate data storage.
<property>
<name>fs.s3a.multipart.size</name>
<value>104857600</value>
<description>
How big (in bytes) to split upload or copy operations up into.
<value>100M</value>
<description>How big (in bytes) to split upload or copy operations up into.
A suffix from the set {K,M,G,T,P} may be used to scale the numeric value.
</description>
</property>

13
hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/ITestS3AConfiguration.java
View File

@ -380,7 +380,7 @@ public class ITestS3AConfiguration {
byte[] file = ContractTestUtils.toAsciiByteArray("test file");
ContractTestUtils.writeAndRead(fs,
new Path("/path/style/access/testFile"), file, file.length,
conf.getInt(Constants.FS_S3A_BLOCK_SIZE, file.length), false, true);
(int) conf.getLongBytes(Constants.FS_S3A_BLOCK_SIZE, file.length), false, true);
} catch (final AWSS3IOException e) {
LOG.error("Caught exception: ", e);
// Catch/pass standard path style access behaviour when live bucket
@ -451,6 +451,17 @@ public class ITestS3AConfiguration {
tmp1.getParent(), tmp2.getParent());
}
@Test
public void testReadAheadRange() throws Exception {
conf = new Configuration();
conf.set(Constants.READAHEAD_RANGE, "300K");
fs = S3ATestUtils.createTestFileSystem(conf);
assertNotNull(fs);
long readAheadRange = fs.getReadAheadRange();
assertNotNull(readAheadRange);
assertEquals("Read Ahead Range Incorrect.", 300 * 1024, readAheadRange);
}
@Test
public void testUsernameFromUGI() throws Throwable {
final String alice = "alice";