DStream.groupByKey(numPartitions=None)[source]#

Return a new DStream by applying groupByKey on each RDD.