[
https://issues.apache.org/jira/browse/HADOOP-17280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17200549#comment-17200549
]
Takanobu Asanuma commented on HADOOP-17280:
-------------------------------------------
Thanks for your suggestion, [~LiJinglun].
Actually, I considered this implementation when implementing HADOOP-17165. I
think there are pros and cons for not calculating the RPC costs of
service-users. The cons are the following:
* It becomes difficult to promote higher priority queues for users who are in
lower priority queues, and they will be penalized more than expected.
* The RPC costs of service-users will no longer appear in CallValume (Raw
Total incoming Call Volume) metrics.
Based on the above, I thought it would be easier to understand if we treated
service users the same way as other users in the highest priority queue, i.e.,
if we also calculated the RPC costs of service users.
> Service-user cost shouldn't be accumulated to totalDecayedCallCost and
> totalRawCallCost.
> ----------------------------------------------------------------------------------------
>
> Key: HADOOP-17280
> URL: https://issues.apache.org/jira/browse/HADOOP-17280
> Project: Hadoop Common
> Issue Type: Improvement
> Reporter: Jinglun
> Assignee: Jinglun
> Priority: Major
> Attachments: HADOOP-17280.001.patch
>
>
> HADOOP-17165 has introduced a very useful feature: service-user. After this
> feature I think we shouldn't add the service-user's cost into
> totalDecayedCallCost and totalRawCallCost anymore. Because it may give all
> the identities the priority 0(Supposing we have a big service-user).
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: common-issues-h...@hadoop.apache.org
