Github user tnachen commented on a diff in the pull request:
https://github.com/apache/spark/pull/3861#discussion_r24648354
--- Diff: docs/running-on-mesos.md ---
@@ -226,6 +226,22 @@ See the [configuration page](configuration.html) for
information on Spark config
The final total amount of memory allocated is the maximum value
between executor memory plus memoryOverhead, and overhead fraction (1.07) plus
the executor memory.
</td>
</tr>
+<tr>
+ <td><code>spark.mesos.shuffle.service.cpu</code></td>
+ <td>1</td>
+ <td>
+ The amount of CPU that the Mesos coarse grained scheduler will request
for launching the shuffle service. The shuffle service is launched on each
slave.
+ This will only apply if shuffle service is enabled and is running
under coarse grained mode.
--- End diff --
Mesos slave doesn't need CPU allocated to it, since it was the one that
offered the CPU in the first place.
This is the amount of CPU that is required for shuffle service itself. So
in the end the total amount of CPU to use is the mesos executor CPU + task CPU
+ shuffle service CPU.
---
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]
