Redshift vs Athena

Overview

What is being compared

Redshift and Athena both run SQL analytics on AWS, one on a cluster you size and one with no cluster at all. The choice depends on query frequency, data volume and whether you want to manage cluster capacity.

• What it is — Managed columnar data warehouse — Serverless interactive query service (Presto)
• Where data lives — On dedicated nodes — In S3, queried directly
• Scale and speed — Petabyte-scale, consistent query performance — Per-query
• Pricing — For the cluster — Per terabyte scanned
• Capacity — You manage the cluster — Nothing to manage
• Fits — Frequent complex queries, sub-second dashboards, distribution and sort keys, data already in Redshift — Ad-hoc S3 exploration, infrequent queries, ETL over raw files, pay-per-query

Load once, or scan every time

Redshift pays the S3 read once, at load time, and answers every later query from its own nodes. Athena reads S3 on every query.

> [!TIP] Both engines side by side, the steps that cut an Athena scan and the choice between them play step by step in Visual Explanation below.

More Comparisons topics

• Kafka vs SQS
• Kinesis vs Kafka
• PrivateLink vs VPC Peering
• Prometheus vs Datadog
• RabbitMQ vs SQS
• RDS vs DynamoDB
• Rolling vs Blue-Green Deployment
• Route 53 Routing Policies
• S3 vs EFS vs EBS
• Spinnaker vs Argo CD
• SQL vs NoSQL
• SQS vs EventBridge