hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Konstantin Shvachko (JIRA)" <j...@apache.org>
Subject [jira] Created: (HADOOP-193) DFS i/o benchmark.
Date Thu, 04 May 2006 01:59:17 GMT
DFS i/o benchmark.

         Key: HADOOP-193
         URL: http://issues.apache.org/jira/browse/HADOOP-193
     Project: Hadoop
        Type: Test

  Components: fs  
    Reporter: Konstantin Shvachko
 Attachments: IOBenchmark.patch

DFS i/o benchmark is a map-reduce based test that measures performance of the cluster for
reads and writes.
This is an evolved version of HADOOP-72, and HADOOP-95 test.

This test writes into or reads from a specified number of files.
File size is specified as a parameter to the test.
Each file is processed in a separate map task.
The unique reducer then collects stats.
Finally, the following information is displayed

# read or write test
# date and time the test finished
# number of files processed
# total number of bytes processed
# throughput in mb/sec (total number of bytes / sum of processing times)
# average i/o rate in mb/sec per file
# standard i/o rate deviation

I  included the test into the AllTestDriver.

This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
For more information on JIRA, see:

View raw message