Using spark locally

Using spark locally.md

Based on https://medium.com/sicara/get-started-pyspark-jupyter-guide-tutorial-ae2fe84f594f

1. Requirements

• Java (e.g. via VS Code) - e.g. Oracle_JDK-22
• Spark (prebuilt via http://spark.apache.org/downloads.html) - e.g. spark-3.5.1-bin-hadoop3
• PySpark (via pip)

2. Add the java bin and spark bin paths to the PATH variable

• C:\Users\matthew\repos\Oracle_JDK-22\bin
• C:\Users\matthew\repos\spark-3.5.1-bin-hadoop3\bin

3. Add environment variables pointing to spark-hadoop, java, and python

• HADOOP_HOME = C:\Users\matthew\repos\spark-3.5.1-bin-hadoop3
• SPARK_HOME = C:\Users\matthew\repos\spark-3.5.1-bin-hadoop3
• JAVA_HOME = C:\Users\matthew\repos\Oracle_JDK-22
• PYSPARK_DRIVER_PYTHON = C:\Users\matthew\Anaconda3\envs\main\python.exe
• PYSPARK_PYTHON = C:\Users\matthew\Anaconda3\envs\main\python.exe

4. Run some test Python code

import pyspark
import random

sc = pyspark.SparkContext(appName="Pi")

num_samples = 100000000

def inside(p):     
  x, y = random.random(), random.random()
  return x*x + y*y < 1

count = sc.parallelize(range(0, num_samples)).filter(inside).count()

pi = 4 * count / num_samples

print(pi)