DataStax News: Astra Streaming now GA with new built-in support for Kafka and RabbitMQ. Read the press release.

Cassandra Total Cost of Ownership Study

Analyst firm, GigaOm, has conducted a comprehensive new study comparing serverless Cassandra with self-managed Cassandra.

Get The Report

Learn Without Fear

We’re excited to share a new learning experience for both new and experienced Cassandra users now at datastax.com/dev.

Try For Free Schedule Demo

Cassandra Data Modeling

Want to use Cassandra successfully? Your data model may be the most important factor! While Cassandra Query Language (CQL) looks like SQL, there are some key differences. Become aware of these differences so you can build a scalable data model.

Get Started

Introduction

You’re using Cassandra because you want your data access to be fast and scalable. The secret to Cassandra’s fast data access is an optimized storage mechanism, which you control with the Primary Key. The primary key, and its components, tells Cassandra how to find your data quickly.

Next Concept: Primary Key

Primary Key

A Cassandra primary key uniquely identifies a row within a Cassandra table. This primary key consists of two parts: a partition key and optional clustering columns. Each of these two parts serve different and specific purposes.

Next Concept: Partition Key

Partition Key

The partition key portion of the primary key consists of one or more columns. Cassandra concatenates all values from the partition key columns and uses the result to locate quickly a partition within the cluster.

Next Concept: Partition

Partition

A partition is a set of rows (a relatively small subset of the table) that shares the same partition key. The partition is a physical unit of access, which means Cassandra will fetch all rows in a partition at the same time — very quickly. You can think of partitions as the results of pre-computed queries.

Next Concept: Clustering Columns

Clustering Columns

Within a partition, Cassandra sorts the rows using the values of the clustering columns. Therefore, during a query, Cassandra can use the clustering column values to search the partition quickly for a specific row within the partition.