Basic Queries of Cloudberry Database
This document introduce the basic queries of Cloudberry Database.
Cloudberry Database is a high-performance, highly parallel data warehouse developed based on PostgreSQL and Greenplum. Here are some examples of the basic query syntax.
-
SELECT
: Used to retrieve data from databases & tables.SELECT * FROM employees; -- Queries all data in the employees table.
-
Conditional query (
WHERE
): Used to filter result sets based on certain conditions.SELECT * FROM employees WHERE salary > 50000; -- Queries employee information with salary exceeding 50,000.
-
ORDER BY
: Used to sort query results by one or more columns.SELECT * FROM employees ORDER BY salary DESC; -- Sorts employee information in descending order by salary.
-
Aggregation functions: such as
COUNT
,SUM
,AVG
,MAX
,MIN
, used for calculating statistics from datasets.SELECT AVG(salary) FROM employees; -- Calculates the average salary of employees.
-
GROUP BY
: Used in conjunction with aggregation functions to group result sets.SELECT department, COUNT(*) FROM employees GROUP BY department; -- Counts the number of employees by department.
-
Limit the number of results (
LIMIT
): used to limit the number of rows returned by the query result.SELECT * FROM employees LIMIT 10; -- Only queries the information of the first 10 employees.
-
Join query (
JOIN
): used to combine data from two or more tables based on related columns.SELECT employees.name, departments.name
FROM employees
JOIN departments ON employees.department_id = departments.id; -- Queries employees and their corresponding department names. -
Subquery: Nested queries in another SQL query.
SELECT name FROM employees
WHERE department_id IN (SELECT id FROM departments WHERE location = 'New York'); -- Queries all employees working in New York.
The above is just a brief overview of the basic query syntax in Cloudberry Database. Cloudberry Database also provides more advanced queries and functions to help developers perform complex data operations and analyses.