You are viewing a plain text version of this content. The canonical link for it is here.

Posted to common-user@hadoop.apache.org by Matias Silva <MS...@specificmedia.com> on 2010/06/18 01:11:13 UTC

Job Metrics Cache Location

Hi Everyone, thanks for your time and knowledge.  I'm looking for where Hadoop keeps
cached information on Job Counters, FileSystemCounters, and Map-Reduce Frameworks
metrics.  Here's an example.


Counter

Map

Reduce

Total

Job Counters

Rack-local map tasks

0

0

1

Launched map tasks

0

0

1



FileSystemCounters

HDFS_BYTES_READ

5,420,676

0

5,420,676

HDFS_BYTES_WRITTEN

155,445,341

0

155,445,341



Map-Reduce Framework

Map input records

515,486

0

515,486

Spilled Records

0

0

0



Map input bytes

5,373,665

0

5,373,665



Map output records

515,486

0

515,486





I'm trying to calculate io.sort.record.percent by greping the cache files for finished
jobs.

Thanks,
Matt