From 08cae491497f1a81c89ad662bb6442bc43d8d6d7 Mon Sep 17 00:00:00 2001
From: Sela <ansela@paypal.com>
Date: Sun, 5 Mar 2017 23:37:08 +0200
Subject: [PATCH] Remove cache of MapWithStateDStream on read.

---
 .../org/apache/beam/runners/spark/io/SparkUnboundedSource.java  | 2 --
 1 file changed, 2 deletions(-)

diff --git a/runners/spark/src/main/java/org/apache/beam/runners/spark/io/SparkUnboundedSource.java b/runners/spark/src/main/java/org/apache/beam/runners/spark/io/SparkUnboundedSource.java
index 354461fac263..e5bbaf185e05 100644
--- a/runners/spark/src/main/java/org/apache/beam/runners/spark/io/SparkUnboundedSource.java
+++ b/runners/spark/src/main/java/org/apache/beam/runners/spark/io/SparkUnboundedSource.java
@@ -94,8 +94,6 @@ public static <T, CheckpointMarkT extends CheckpointMark> UnboundedDataset<T> re
 
     // set checkpoint duration for read stream, if set.
     checkpointStream(mapWithStateDStream, options);
-    // cache since checkpointing is less frequent.
-    mapWithStateDStream.cache();
 
     // report the number of input elements for this InputDStream to the InputInfoTracker.
     int id = inputDStream.inputDStream().id();