hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Doug Cutting (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-331) map outputs should be written to a single output file with an index
Date Mon, 16 Oct 2006 22:34:40 GMT
    [ http://issues.apache.org/jira/browse/HADOOP-331?page=comments#action_12442732 ] 
            
Doug Cutting commented on HADOOP-331:
-------------------------------------

> sort each bucket and reduce the number of comparisons and swaps done considerably

In memory, we only need to sort pointers to buffered items, so there only needs to be a single
buffer with the serialized items.  Still, I agree that we can minimize comparisons & swaps
by bucketing these pointers, then sorting each buffer prior to dumping the data it points
to.  Is that what you meant, Sameer?

Each spill can result in output of the same format as the final output: a <key,value>*
file sorted by <part,key> and a <part,offset>* file that indicates the start of
each part.  So if there's only one spill then no merging is required.  Unfortunately, since
'part' is implicit, we cannot use the existing SequenceFile merge code.


> map outputs should be written to a single output file with an index
> -------------------------------------------------------------------
>
>                 Key: HADOOP-331
>                 URL: http://issues.apache.org/jira/browse/HADOOP-331
>             Project: Hadoop
>          Issue Type: Improvement
>          Components: mapred
>    Affects Versions: 0.3.2
>            Reporter: eric baldeschwieler
>         Assigned To: Devaraj Das
>
> The current strategy of writing a file per target map is consuming a lot of unused buffer
space (causing out of memory crashes) and puts a lot of burden on the FS (many opens, inodes
used, etc).  
> I propose that we write a single file containing all output and also write an index file
IDing which byte range in the file goes to each reduce.  This will remove the issue of buffer
waste, address scaling issues with number of open files and generally set us up better for
scaling.  It will also have advantages with very small inputs, since the buffer cache will
reduce the number of seeks needed and the data serving node can open a single file and just
keep it open rather than needing to do directory and open ops on every request.
> The only issue I see is that in cases where the task output is substantiallyu larger
than its input, we may need to spill multiple times.  In this case, we can do a merge after
all spills are complete (or during the final spill).

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message