[
https://issues.apache.org/jira/browse/MAPREDUCE-931?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dick King updated MAPREDUCE-931:
--------------------------------
Attachment: MAPREDUCE-931--2009-09-16--1633.patch
This patch fixes a small bug in {{CDFRandomGenerator}}, and also makes
{{ZombieJob}} use a {{CDFPiecewiseLinearRandomGenerator}} to do its
interpolation from a CDF rather than the ad hoc code that was there. This is
an important code cleanup because the old code built new tables for each CDF,
and more importantly because it makes plugging in a new kind of interpolator in
the future a one-line change.
We did not add tests. The results are non-deterministic. The patch testing
tool will complain that there is no new test case. I did check some output
manually and it looked plausible.
> rumen should use its own interpolation classes to create runtimes for
> simulated tasks
> -------------------------------------------------------------------------------------
>
> Key: MAPREDUCE-931
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-931
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Reporter: Dick King
> Assignee: Dick King
> Priority: Minor
> Attachments: MAPREDUCE-931--2009-09-16--1633.patch, patch-931-b.patch
>
>
> Currently, when a simulator or benchmark is running and simulating hadoop
> jobs using rumen data, and rumen's runtime system is used to get execution
> times for the tasks in the simulated jobs, rumen would use some ad hoc code,
> despite the fact that rumen has a perfectly good interpolation framework to
> generate random variables that fit discrete CDFs.
> We should use the interpolation framework.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
