Github user JackYangzg commented on the issue:
https://github.com/apache/spark/pull/18435
@jerryshao I agree with you point, it enough to use o.cores / CPUS_PER_TASK
to insteadãlike the example you list, the left one core is not enough to
allocate for a task.
In fact, to deal with this
`val tasks = shuffledOffers.map(o => new
ArrayBuffer[TaskDescription](o.cores / CPUS_PER_TASK))`
is easy to understand the way how to allocate offers for the task .
@srowen
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
