Word count program using RDDs

Input file
sample.txt:
-----------
spark is  bigdata technology
hadoop is bigdata technology
spark and hadoop are bigdata technologies



Word count program using RDDs:
---------------------------------
scala> val r1 = sc.textFile("e:\\vow\\sample.txt")
r1: org.apache.spark.rdd.RDD[String] = e:\vow\sample.txt MapPartitionsRDD[3] at textFile at <console>:24

scala> r1.count
res1: Long = 3

scala> r1.collect.foreach(println)
spark is  bigdata technology
hadoop is bigdata technology
spark and hadoop are bigdata technologies

scala> val r2 = r1.map(x => x.split(" "))
r2: org.apache.spark.rdd.RDD[Array[String]] = MapPartitionsRDD[4] at map at <console>:25

scala> r2.take(3)
res4: Array[Array[String]] = Array(Array(spark, is, "", bigdata, technology), Array(hadoop, is, bigdata, technology), Array(spark, and, hadoop, are, b
igdata, technologies))

scala> val r3 = r2.flatMap(x => x)
r3: org.apache.spark.rdd.RDD[String] = MapPartitionsRDD[5] at flatMap at <console>:25

scala> r3.collect
res5: Array[String] = Array(spark, is, "", bigdata, technology, hadoop, is, bigdata, technology, spark, and, hadoop, are, bigdata, technologies)

scala> val r4 = r3.map(x => (x,1))
r4: org.apache.spark.rdd.RDD[(String, Int)] = MapPartitionsRDD[7] at map at <console>:25

scala> r4.collect
res7: Array[(String, Int)] = Array((spark,1), (is,1), ("",1), (bigdata,1), (technology,1), (hadoop,1), (is,1), (bigda
 (and,1), (hadoop,1), (are,1), (bigdata,1), (technologies,1))

scala> val r5  = r4.reduceByKey( (x,y) => (x+y))
r5: org.apache.spark.rdd.RDD[(String, Int)] = ShuffledRDD[9] at reduceByKey at <console>:25

scala> r5.collect.foreach(println)
(technology,2)
(are,1)
(is,2)
(,1)
(bigdata,3)
(technologies,1)
(spark,2)
(hadoop,2)
(and,1)