Github user andrewor14 commented on a diff in the pull request:
https://github.com/apache/spark/pull/12154#discussion_r58496003
--- Diff:
streaming/src/main/scala/org/apache/spark/streaming/scheduler/ExecutorAllocationManager.scala
---
@@ -0,0 +1,217 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+
+package org.apache.spark.streaming.scheduler
+
+import scala.util.Random
+
+import org.apache.spark.{ExecutorAllocationClient, SparkConf}
+import org.apache.spark.internal.Logging
+import org.apache.spark.streaming.util.RecurringTimer
+import org.apache.spark.util.{Clock, Utils}
+
+/**
+ * Class that manages executor allocated to a StreamingContext, and
dynamically request or kill
+ * executors based on the statistics of the streaming computation. At a
high level, the policy is:
+ * - Use StreamingListener interface get batch processing times of
completed batches
+ * - Periodically take the average batch completion times and compare with
the batch interval
+ * - If (avg. proc. time / batch interval) >= scaling up ratio, then
request more executors
--- End diff --
can you expand on this java doc to comment on how this is different from
the core dynamic allocation? We should mention that this intends to stabilize
the system over time gradually by requesting and killing executors 1 at a time.
Bonus points if you add a paragraph on how well this works with backpressure.
---
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]
