You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@tez.apache.org by "Hitesh Shah (JIRA)" <ji...@apache.org> on 2014/10/17 00:33:33 UTC

[jira] [Commented] (TEZ-1674) Rename parameters related to counters / memory distribution

    [ https://issues.apache.org/jira/browse/TEZ-1674?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14174374#comment-14174374 ] 

Hitesh Shah commented on TEZ-1674:
----------------------------------

{code}
     mrParamToDAGParamMap.put(MRJobConfig.COUNTERS_MAX_KEY,
-      TezConfiguration.TEZ_AM_COUNTERS_MAX_KEYS);
+      TezConfiguration.TEZ_COUNTERS_MAX);
     mrParamToDAGParamMap.put(MRJobConfig.COUNTER_GROUP_NAME_MAX_KEY,
-      TezConfiguration.TEZ_AM_COUNTERS_GROUP_NAME_MAX_KEYS);
+      TezConfiguration.TEZ_COUNTERS_GROUP_NAME_MAX_LENGTH);
     mrParamToDAGParamMap.put(MRJobConfig.COUNTER_GROUP_NAME_MAX_KEY,
-      TezConfiguration.TEZ_AM_COUNTERS_NAME_MAX_KEYS);
+      TezConfiguration.TEZ_COUNTERS_COUNTER_NAME_MAX_LENGTH);
     mrParamToDAGParamMap.put(MRJobConfig.COUNTER_GROUPS_MAX_KEY,
-      TezConfiguration.TEZ_AM_COUNTERS_GROUPS_MAX_KEYS);
+      TezConfiguration.TEZ_COUNTERS_MAX_GROUPS);
   }
{code}

I see "mrParamToDAGParamMap.put(MRJobConfig.COUNTER_GROUP_NAME_MAX_KEY" show up twice. Is that a bug from before? 

Rest looks fine.




> Rename parameters related to counters / memory distribution
> -----------------------------------------------------------
>
>                 Key: TEZ-1674
>                 URL: https://issues.apache.org/jira/browse/TEZ-1674
>             Project: Apache Tez
>          Issue Type: Improvement
>            Reporter: Siddharth Seth
>            Assignee: Siddharth Seth
>            Priority: Critical
>         Attachments: TEZ-1674.1.txt
>
>
> The parameter names don't make a lot of sense right now, or duplicate information.
> Previous name   Current name after the 0.5 refactor.
> tez.runtime.job.counters.max  tez.am.counters.max.keys
> tez.runtime.job.counters.group.name.max tez.am.counters.group-name.max.keys
> tez.runtime.job.counters.counter.name.max tez.am.counters.name.max.keys
> tez.runtime.job.counters.groups.max tez.am.counters.groups.max.keys
> A lot of these are not related to keys
> tez.task.scale.memory.allocator.class tez.task.scale.task.memory.allocator.class
> tez.task.scale.memory.reserve-fraction  tez.task.scale.task.memory.reserve-fraction
> tez.task.scale.memory.additional.reservation.fraction.per-io  tez.task.scale.task.memory.additional-reservation.fraction.per-io
> tez.task.scale.memory.additional reservation.fraction.max tez.task.scale.task.memory.additional-reservation.fraction.max
> tez.task.initial.memory.scale.ratios  tez.task.scale.task.memory.ratios
> task is duplicated
> Also
> tez.task.max-events-per-heartbeat.max - repeats max
> I'd prefer to have made these changes directly, but after talking to [~hitesh], will add deprecation - and target removal in 0.7



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)