alluxio.inode

Class InodeBenchRead

java.lang.Object

alluxio.inode.InodeBenchRead

public class InodeBenchRead
extends Object

This benchmark measures the time it takes to read inodes from a tree like structure, this includes locking and traversing the read path. The following parameters can be varied: mSingleFile - if true the operation measured is to traverse the tree and return a single inode. If false the operation measured is to traverse the tree and list the files in the reached directory. mDepth - the number of levels in the inode tree mFileCount - the number of inodes created at each depth mUseZipf - if true depths and inodes to be read will be chosen according to a Zipfian distribution. This means that directories with shallow depth will be more likely to be chosen, and files with larger ids are more likely to be chosen (i.e. those written later). mType - the type of inode storage to use mRocksConfig - see RocksBenchConfig

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	InodeBenchRead.Db
static class	InodeBenchRead.ThreadState

Constructor Summary

Constructors

Constructor and Description
InodeBenchRead()

Method Summary

Modifier and Type	Method and Description
static void	main(String[] args)
void	testMethod(InodeBenchRead.Db db, InodeBenchRead.ThreadState ts, org.openjdk.jmh.infra.Blackhole bh)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

InodeBenchRead

public InodeBenchRead()

Method Detail

testMethod

public void testMethod(InodeBenchRead.Db db,
                       InodeBenchRead.ThreadState ts,
                       org.openjdk.jmh.infra.Blackhole bh)
                throws Exception

Throws:

Exception

main

public static void main(String[] args)
                 throws org.openjdk.jmh.runner.RunnerException

Throws:

org.openjdk.jmh.runner.RunnerException