[
https://issues.apache.org/jira/browse/HADOOP-14600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16264406#comment-16264406
]
Ohad Raviv commented on HADOOP-14600:
-------------------------------------
Hi,
We also encountered a similar problem as SPARK-21137, when we debugged it we
saw that all the time got consumed in the initilazation of LocatedFileStatus
when in FileInputFormat tries to get:
{code}
RemoteIterator<LocatedFileStatus> iter =
fs.listLocatedStatus(globStat.getPath());
{code}
However, non of the "special" methods of LocatedFileStatus are used in the
loop, so changing it to:
{code}
for (FileStatus globStat: matches) {
if (globStat.isDirectory()) {
RemoteIterator<FileStatus> iter =
fs.listStatusIterator(globStat.getPath());
while (iter.hasNext()) {
FileStatus stat = iter.next();
if (inputFilter.accept(stat.getPath())) {
if (recursive && stat.isDirectory()) {
addInputPathRecursively(result, fs, stat.getPath(),
inputFilter);
} else {
result.add(stat);
}
}
}
} else {
result.add(globStat);
}
}
{code}
actually solved all the problem.
I know that it is not the only place that needs to change,
but I would like to know what do you think about that?
thanks,
Ohad.
> LocatedFileStatus constructor forces RawLocalFS to exec a process to get the
> permissions
> ----------------------------------------------------------------------------------------
>
> Key: HADOOP-14600
> URL: https://issues.apache.org/jira/browse/HADOOP-14600
> Project: Hadoop Common
> Issue Type: Bug
> Components: fs
> Affects Versions: 2.7.3
> Environment: file:// in a dir with many files
> Reporter: Steve Loughran
> Assignee: Ping Liu
> Attachments: HADOOP-14600.001.patch, HADOOP-14600.002.patch,
> HADOOP-14600.003.patch, HADOOP-14600.004.patch, HADOOP-14600.005.patch,
> HADOOP-14600.006.patch, HADOOP-14600.007.patch, HADOOP-14600.008.patch,
> HADOOP-14600.009.patch, TestRawLocalFileSystemContract.java
>
>
> Reported in SPARK-21137. a {{FileSystem.listStatus}} call really craws
> against the local FS, because {{FileStatus.getPemissions}} call forces
> {{DeprecatedRawLocalFileStatus}} tp spawn a process to read the real UGI
> values.
> That is: for every other FS, what's a field lookup or even a no-op, on the
> local FS it's a process exec/spawn, with all the costs. This gets expensive
> if you have many files.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
