[
https://issues.apache.org/jira/browse/YARN-6163?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15862036#comment-15862036
]
ASF GitHub Bot commented on YARN-6163:
--------------------------------------
Github user templedf commented on a diff in the pull request:
https://github.com/apache/hadoop/pull/192#discussion_r100633456
--- Diff:
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FairSchedulerConfiguration.java
---
@@ -114,12 +114,24 @@
protected static final String PREEMPTION_THRESHOLD =
CONF_PREFIX + "preemption.cluster-utilization-threshold";
protected static final float DEFAULT_PREEMPTION_THRESHOLD = 0.8f;
-
- protected static final String PREEMPTION_INTERVAL = CONF_PREFIX +
"preemptionInterval";
- protected static final int DEFAULT_PREEMPTION_INTERVAL = 5000;
+
protected static final String WAIT_TIME_BEFORE_KILL = CONF_PREFIX +
"waitTimeBeforeKill";
protected static final int DEFAULT_WAIT_TIME_BEFORE_KILL = 15000;
+ /**
+ * Configurable delay before an app's starvation is considered after it
is
+ * identified. This is to give the scheduler enough time to
+ * allocate containers post preemption. This delay is added to the
+ * {@link #WAIT_TIME_BEFORE_KILL} and enough heartbeats.
+ *
+ * This is intended as a backdoor on production clusters, and hence
+ * intentionally not documented.
+ */
+ protected static final String WAIT_TIME_BEFORE_NEXT_STARVATION_CHECK =
--- End diff --
The name and description should include the units.
> FS Preemption is a trickle for severely starved applications
> ------------------------------------------------------------
>
> Key: YARN-6163
> URL: https://issues.apache.org/jira/browse/YARN-6163
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: fairscheduler
> Affects Versions: 2.9.0
> Reporter: Karthik Kambatla
> Assignee: Karthik Kambatla
> Attachments: yarn-6163-1.patch
>
>
> With current logic, only one RR is considered per each instance of marking an
> application starved. This marking happens only on the update call that runs
> every 500ms. Due to this, an application that is severely starved takes
> forever to reach fairshare based on preemptions.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
