- 创建文件:
在Linux文件夹下,进如要文件的目录:输入vi test.txt
按i输入 文本内容,输入完毕按esc,按:wq 保存退出
然后在命令行执行下面的命令:
hadoop fs –mkdir –p /Hadoop/Input (在HDFS创建目录)
hadoop fs –put test.txt /Hadoop/Input (将test.txt文件上传到HDFS)
hadoop fs –ls /Hadoop/Input (查看上传的文件)
hadoop fs –text /Hadoop/Input/test.txt (查看文件的内容)
启动spark
spark-shell
输入下面的语句:
val file=sc.textFile(“/Hadoop/Input/test.txt”)
val rdd = file.flatMap(line => line.split(“ ”)).map(word => (word,1)).reduceByKey(_+_)
val maps = rdd.collect()
for (i <- 0 to maps.length-1) {
println(maps(i))
}