[ 
https://issues.apache.org/jira/browse/HADOOP-492?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12473474
 ] 

Doug Cutting commented on HADOOP-492:
-------------------------------------

That sounds like a great plan.

> Do we need some sort of counter naming convention to prevent future conflicts 
> between framework-maintained counters and user-defined counters?

We could perhaps piggyback of  Java's naming system by changing the Reporter 
method to be:

   void incrCounter(Enum key, long amount);

Then, internally, we can convert the key to a String with something like:

   String name = key.getDeclaringClass().getName()+"#"+key.toString();

This serves two purposes: keys are checked at compile time (since they have to 
be defined with enums) and they're also package-qualified.

In the web ui, it would be great if all counters, both user and system defined, 
were displayed in various forms: raw totals, total rates (counts/second), and 
per-task averages (average count/task, average rate/task).


> Global counters
> ---------------
>
>                 Key: HADOOP-492
>                 URL: https://issues.apache.org/jira/browse/HADOOP-492
>             Project: Hadoop
>          Issue Type: New Feature
>          Components: mapred
>            Reporter: arkady borkovsky
>         Assigned To: David Bowen
>
> It would be nice to have map / reduce job keep aggregated counts for 
> arbitrary events occuring in its tasks -- the numer of records processed, the 
> numer of exceptions of a specific type, the number of sentences in passive 
> voice, whatever the jobs finds useful.
> This can be implemented by tasks periodically sending <name, value> pairs to 
> the jobtracker (in some implementations such messages are piggy-backed on the 
> heartbeats), so that the job tracker stores all the latests values from each 
> task and aggregates them on a request.  It should also make the aggregated 
> values available at the job end.  The value for a task would be flushed when 
> the task fails.
> #491 and #490 may be related to this one.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to