[ 
https://issues.apache.org/jira/browse/SPARK-27760?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Thomas Graves updated SPARK-27760:
----------------------------------
    Description: 
For the Spark resources, we created the config

spark.\{driver/executor}.resource.\{resourceName}.count

I think we should change .count to be .amount. That more easily allows users to 
specify things with suffix like memory in a single config and they can combine 
the value and unit. Without this they would have to specify 2 separate configs 
(like .count and .unit) which seems more of a hassle for the user.

Note the yarn configs for resources use amount:  
spark.yarn.\{executor/driver/am}.resource=<some amount>, where the <some amont> 
is value and unit together. I think that makes a lot of sense. Filed a separate 
Jira to add .amount to the yarn configs as well.

  was:
For the Spark resources, we created the config

spark.\{driver/executor}.resource.\{resourceName}.count

I think we should change .count to be .amount. That more easily allows users to 
specify things with suffix like memory in a single config and they can combine 
the value and unit. Without this they would have to specify 2 separate configs 
which seems more of a hassle for the user.

Note the yarn configs for resources use amount:  
spark.yarn.\{executor/driver/am}.resource=<some amount>, where the <some amont> 
is value and unit together. I think that makes a lot of sense. Filed a separate 
Jira to add .amount to the yarn configs as well.


> Spark resources - user configs change .count to be .amount
> ----------------------------------------------------------
>
>                 Key: SPARK-27760
>                 URL: https://issues.apache.org/jira/browse/SPARK-27760
>             Project: Spark
>          Issue Type: Story
>          Components: Spark Core
>    Affects Versions: 3.0.0
>            Reporter: Thomas Graves
>            Assignee: Thomas Graves
>            Priority: Major
>
> For the Spark resources, we created the config
> spark.\{driver/executor}.resource.\{resourceName}.count
> I think we should change .count to be .amount. That more easily allows users 
> to specify things with suffix like memory in a single config and they can 
> combine the value and unit. Without this they would have to specify 2 
> separate configs (like .count and .unit) which seems more of a hassle for the 
> user.
> Note the yarn configs for resources use amount:  
> spark.yarn.\{executor/driver/am}.resource=<some amount>, where the <some 
> amont> is value and unit together. I think that makes a lot of sense. Filed a 
> separate Jira to add .amount to the yarn configs as well.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org

Reply via email to