You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2022/03/17 15:40:27 UTC

[GitHub] [spark] Yikun commented on a change in pull request #35504: [SPARK-38194][YARN][MESOS][K8S] Make memory overhead factor configurable

Yikun commented on a change in pull request #35504:
URL: https://github.com/apache/spark/pull/35504#discussion_r829192039



##########
File path: resource-managers/kubernetes/core/src/main/scala/org/apache/spark/deploy/k8s/features/BasicExecutorFeatureStep.scala
##########
@@ -59,11 +59,16 @@ private[spark] class BasicExecutorFeatureStep(
   private val isDefaultProfile = resourceProfile.id == ResourceProfile.DEFAULT_RESOURCE_PROFILE_ID
   private val isPythonApp = kubernetesConf.get(APP_RESOURCE_TYPE) == Some(APP_RESOURCE_TYPE_PYTHON)
   private val disableConfigMap = kubernetesConf.get(KUBERNETES_EXECUTOR_DISABLE_CONFIGMAP)
+  private val memoryOverheadFactor = if (kubernetesConf.contains(EXECUTOR_MEMORY_OVERHEAD_FACTOR)) {
+    kubernetesConf.get(EXECUTOR_MEMORY_OVERHEAD_FACTOR)
+  } else {
+    kubernetesConf.get(MEMORY_OVERHEAD_FACTOR)
+  }

Review comment:
       The reason should be in here, before kubernetesConf.get(MEMORY_OVERHEAD_FACTOR) was used as default factor, it's `0.4` according my real debug watch on IT `Run PySpark on simple pi.py example`.
   
   But current EXECUTOR_MEMORY_OVERHEAD_FACTOR has more priority than so MEMORY_OVERHEAD_FACTOR is be overrited. (so 0.1 by default). So that the default behavior changed.
   
   But I haven't found why `kubernetesConf.get(MEMORY_OVERHEAD_FACTOR)` is `0.4` yet, I couldn't find a code in IT to set this explictly.
   
   cc @Kimahriman @tgravescs 




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org