Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
MapReduce, mail # user - temporary file locations for YARN applications


Copy link to this message
-
RE: temporary file locations for YARN applications
John Lilley 2013-10-21, 22:49
Thanks, sounds like LOCAL_DIR_ENV is the way to go.
john

-----Original Message-----
From: Harsh J [mailto:[EMAIL PROTECTED]]
Sent: Monday, October 21, 2013 12:11 PM
To: <[EMAIL PROTECTED]>
Subject: Re: temporary file locations for YARN applications

The dirs in that env-var are app-specific and are for the app's user to utilize. You shouldn't have any permission issues working within them.

The LocalDirAllocator is still somewhat MR-bound but you can still be able to make it work by giving it a config with the values it needs.

On Mon, Oct 21, 2013 at 8:49 PM, John Lilley <[EMAIL PROTECTED]> wrote:
> Thanks again.  This gives me a lot of options; we will see what works.
>
> Do you know if there are any permissions issues if we directly access the folders of LOCAL_DIR_ENV?
>
> Regarding LocalDirAllocator, I see its constructor: LocalDirAllocator(String contextCfgItemName) and a note mentioning that an example of this item is "mapred.local.dir".  Is that the correct usage, or is there something YARN-generic?
>
> Cheers,
> john
>
> -----Original Message-----
> From: Harsh J [mailto:[EMAIL PROTECTED]]
> Sent: Sunday, October 20, 2013 11:58 PM
> To: <[EMAIL PROTECTED]>
> Subject: Re: temporary file locations for YARN applications
>
> Hi,
>
> MR does use multiple disks when spilling. But the work directory is also round-robined to spread I/O.
>
> YARN sets an environment property thats a list (comma separated value)
> of directories (ApplicationConstants.LOCAL_DIR_ENV) your app container
> can together use. Perhaps read it in with
> StringUtils.getTrimmedStrings(System.getenv(ApplicationConstants.LOCAL
> _DIR_ENV)); and then round robin internally over those paths (with
> free space handling)?
>
> Perhaps you can even reuse the org.apache.hadoop.fs.LocalDirAllocator
> class; which is what MR uses. Its not been declared publicly stable though, but we can do that over a JIRA.
>
> On Mon, Oct 21, 2013 at 2:05 AM, John Lilley <[EMAIL PROTECTED]> wrote:
>> Harsh, thanks for the quick response.  These files don't need to be on the DFS (although we use that too).  These are local files used during sorting, joining, transitive closure.
>>
>> The task-relative folder might be good enough, but our app *can* make use of multiple temp folders if they are available.  Our YARN app can be fairly I/O intensive; is it possible to allocate more than one temp folder on different physical devices?
>>
>> Or perhaps YARN might help us. Will YARN assign tasks to CWD folders on different disks so that they do not compete with each other on I/O?
>>
>> For that matter, where does MR allocate the temporary files generated by Mapper output?  Presumably MR has the same I/O parallelism requirements that we do.
>>
>> Thanks
>> John
>>
>>
>> -----Original Message-----
>> From: Harsh J [mailto:[EMAIL PROTECTED]]
>> Sent: Sunday, October 20, 2013 10:49 AM
>> To: <[EMAIL PROTECTED]>
>> Subject: Re: temporary file locations for YARN applications
>>
>> Every container gets its own local work directory (You can use the relative ./) thats auto-cleaned up at the end of the container's life.
>> This is the best place to store the temporary files. This is not something you need custom configuration for.
>>
>> Do the files need to be on a distributed FS or a local one?
>>
>> On Sun, Oct 20, 2013 at 8:54 PM, John Lilley <[EMAIL PROTECTED]> wrote:
>>> We have a pure YARN application (no MapReduce) that has need to
>>> store a significant amount of temporary data.  How can we know the
>>> best location for these files?  How can we ensure that our YARN
>>> tasks have write access to these locations?  Is this something that must be configured outside of YARN?
>>> Thanks,
>>> John
>>
>> --
>> Harsh J
>
>
>
> --
> Harsh J

--
Harsh J