Why Hive enables SQL on Hadoop

Introduction

Hive lets you use SQL, a simple and popular language, to work with big data stored in Hadoop. This makes it easier to analyze large data without learning complex programming.

You want to analyze large datasets stored in Hadoop using familiar SQL queries.

You need to run batch processing jobs on big data without writing Java or Python code.

You want to let non-programmers access and query big data easily.

You need to integrate Hadoop data with existing SQL-based tools and workflows.

Syntax

Hadoop

SELECT column1, column2 FROM table_name WHERE condition;

Hive uses a SQL-like language called HiveQL to query data stored in Hadoop.

You write queries similar to standard SQL, which Hive translates to run on Hadoop.

Examples

Selects names and ages of employees older than 30.

Hadoop

SELECT name, age FROM employees WHERE age > 30;

Counts the number of sales records in the North region.

Hadoop

SELECT COUNT(*) FROM sales WHERE region = 'North';

Creates a table and loads data from a file into Hive.

Hadoop

CREATE TABLE users (id INT, name STRING, email STRING);
LOAD DATA INPATH '/user/data/users.csv' INTO TABLE users;

Sample Program

This code connects to Hive, creates a table, inserts some data, and queries employees older than 30.

Hadoop

from pyhive import hive

# Connect to Hive server
conn = hive.Connection(host='localhost', port=10000, username='user')
cursor = conn.cursor()

# Create a table
cursor.execute('CREATE TABLE IF NOT EXISTS employees (name STRING, age INT)')

# Insert sample data
cursor.execute("INSERT INTO employees VALUES ('Alice', 34), ('Bob', 28), ('Carol', 45)")

# Query data
cursor.execute('SELECT name, age FROM employees WHERE age > 30')
results = cursor.fetchall()

for row in results:
    print(row)

Important Notes

Hive translates SQL queries into MapReduce or other Hadoop jobs behind the scenes.

It is best for batch processing, not real-time queries.

Summary

Hive makes big data in Hadoop accessible using SQL.

It helps users analyze data without deep programming skills.

HiveQL is similar to SQL, making it easy to learn.