Bare metal cluster
Streaming in 1.6 seems to have a memory leak.
Running the same streaming app in Spark 1.5.1 and 1.6, all things equal, 1.6 showed a gradual increasing processing time.
The app is simple: 1 Kafka receiver of tweet stream and 20 executors processing the tweets in 5-second batches.
Spark 1.5.0 handles this smoothly and did not show increasing processing time in the 40-minute test; but 1.6 showed increasing time about 8 minutes into the test. Please see chart here:
I captured heap dumps in two version and did a comparison. I noticed the Byte is using 50X more space in 1.5.1.
Here are some top classes in heap histogram and references.
All Classes (excluding platform)
1.6.0 Streaming 1.5.1 Streaming
Class Instance Count Total Size Class Instance Count Total Size
class [B 8453 3,227,649,599 class [B 5095 62,938,466
class [C 44682 4,255,502 class [C 130482 12,844,182
class java.lang.reflect.Method 9059 1,177,670 class java.lang.String 130171 1,562,052
References by Type References by Type
class [B [0x640039e38] class [B [0x6c020bb08]
Referrers by Type Referrers by Type
Class Count Class Count
java.nio.HeapByteBuffer 3239 sun.security.util.DerInputBuffer 1233
sun.security.util.DerInputBuffer 1233 sun.security.util.ObjectIdentifier 620
sun.security.util.ObjectIdentifier 620 [[B 397
[Ljava.lang.Object; 408 java.lang.reflect.Method 326
The total size by class B is 3GB in 1.5.1 and only 60MB in 1.6.0.
The Java.nio.HeapByteBuffer referencing class did not show up in top in 1.5.1.
I have also placed jstack output for 1.5.1 and 1.6.0 online..you can get them here