Start and Stop Hadoop, Yarn, Spark, and Spark History Server

To start Hadoop, Yarn, Spark, and Spark History Server:

start_hadoop_spark.sh

To stop Hadoop, Yarn, Spark, and Spark History Server:

stop_hadoop_spark.sh

Conda Environment

To create the conda environment:

conda env create -f pyspark_conda_env.tar.gz

To activate the conda environment:

conda activate pyspark_conda_env

Query 1

To run Query 1 with the Dataframe API:

python code/q1_DF.py

To run Query 1 with the SQL API:

python code/q1_SQL.py

Query 2

To run Query 2 with the Dataframe API:

python code/q2_DF.py

To run Query 2 with the RDD API:

python code/q2_RDD.py

Query 3

To change the number of executors change NUM_EXECUTORS

To run Query 3:

python code/q3.py

Query 4

To run Query 4:

python code/q4.py

Useful Commands

Turn off safe mode

hdfs dfsadmin -safemode leave

HDFS Report

hdfs dfsadmin -report

See size information about HDFS filesystem hdfs dfs -du -h /user

Remove Staging Log and event log hdfs dfs -rm -r -skipTrash /user/user/.sparkStaging hdfs dfs -rm -r -skipTrash /spark.eventLog/*