spark 2.4 java8 hello world

download JDK 8, extract and add to .bashrc:

export JAVA_HOME=/home/bonelee/jdk1.8.0_211
export CLASSPATH=.:$JAVA_HOME/lib:$JRE_HOME/lib:$CLASSPATH
export PATH=$JAVA_HOME/bin:$JRE_HOME/bin:$PATH
export JRE_HOME=$JAVA_HOME/jre

download spark, unzip. and run:

./bin/spark-submit ~/src_test/spark_hello.py

spark_hello.py :

from pyspark.context import SparkContext
from pyspark.conf import SparkConf

sc = SparkContext(conf=SparkConf().setAppName("mnist_parallelize"))
text_file = sc.textFile("file:///tmp/test.txt")
counts = text_file.flatMap(lambda line: line.split(" ")) 
             .map(lambda word: (word, 1)) 
             .reduceByKey(lambda a, b: a + b)
print(counts.collect())

/tmp/test.txt

text_file = sc.textFile("hdfs://...")
counts = text_file.flatMap(lambda line: line.split(" ")) 
             .map(lambda word: (word, 1)) 
                          .reduceByKey(lambda a, b: a + b)
                          counts.saveAsTextFile("hdfs://...")

output:

[('100', 1), ('text_file', 1), ('=', 2), ('counts', 1), ('text_file.flatMap(lambda', 1), ('line.split("', 1), ('"))', 1), ('', 65), ('word:', 1), ('(word,', 1), ('1))', 1), ('b:', 1), ('sc.textFile("hdfs://...")', 1), ('line:', 1), ('\', 2), ('.map(lambda', 1), ('.reduceByKey(lambda', 1), ('a,', 1), ('a', 1), ('+', 1), ('b)', 1), ('counts.saveAsTextFile("hdfs://...")', 1)]