I have made some progress - the partitioning is very uneven, and everything
goes to one partition. I see that spark partitions by key, so I tried this:
//partitioning is done like partitionIdx = f(key) % numPartitions
//we use random keys to get even partitioning
val uniform = other_stream.transform(rdd => {
rdd.map({ kv =>
val k = kv._1
val v = kv._2
(UUID.randomUUID().toString, v)
})
})
uniform.foreachRDD(rdd => {
rdd.forEachPartition(partition => {
...
As you can see, I'm using random keys. Even in this case, when running with
2 nodes, i verified that one partition is completely empty, and the other
contains all the records.
What is going wrong with the partitioning here?
--
View this message in context:
http://apache-spark-user-list.1001560.n3.nabble.com/Imbalanced-shuffle-read-tp18648p18790.html
Sent from the Apache Spark User List mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscr...@spark.apache.org
For additional commands, e-mail: user-h...@spark.apache.org
