Amazon Redshift

Explore how Amazon Redshift helps creating high performant data warehouses.

Amazon Redshift is a fully managed data warehousing service that can scale to petabytes of data. It is based on PostgreSQL and can be queried using SQL queries. Amazon Redshift stores data in a columnar format and offers a multi-query engine that optimizes analytic queries.

Press + to interact

How Amazon Redshift works

Redshift uses a distributed architecture with multiple nodes working in parallel to execute queries. This Massively Parallel (MPP) architecture enables Redshift to run complex queries and quickly scale horizontally to process large datasets.

The core component of Redshift is a cluster. A cluster consists of multiple compute nodes and a leader node.

  • We connect to the leader node to interact with the cluster. The leader node generates a query execution plan and aggregates results. It manages the communication between the client and the compute nodes.

  • The compute nodes perform the query and respond to the leader node with the results. Each compute node has its own memory and processing power based on the instance’s size. As the workload increases, we can change the compute instance’s type, number of instances, or both. ...