Contrast vs Redshift
Redshift bills compute as capacity over time: nodes by the hour, or Serverless RPUs by the second with a 60-second minimum. Contrast charges an agent's short queries for the bytes they read. And on Contrast the Iceberg table is the warehouse table.
Updated November 2026
At a glance
- $10
- per TB scanned
- 100%
- tables in Apache Iceberg
- 0
- MCP servers to run
Figures from the Contrast pricing page and docs.
How you're billed
List prices as each vendor publishes them, checked November 2026.
Contrast
- Billed on
- Data scanned and storage
- Price
- $10 per TB scanned
Plus $0.05 per GB-month stored. 50 GB scanned and 10 GB stored free each month.
Redshift
- Billed on
- Capacity over time
- Price
- Nodes from $0.543 an hour
Serverless RPUs from $1.50 an hour, with a 60-second minimum.
Still sizing clusters and RPUs before an agent can ask a question?
Agents send lots of short queries while they work. Here's how each side handles that.
Contrast
Every synced table is Apache Iceberg, so the lake and the warehouse are the same files. On Enterprise they can stay in your own bucket.
Redshift
Native tables live in Redshift Managed Storage, billed apart from compute. Iceberg tables in Glue need an external schema.
Before an agent's first query
On Redshift, the workgroup bills RPUs per second with a 60-second minimum, and the MCP server runs on your machine with your AWS credentials. Contrast's MCP server is hosted, and queries bill by data scanned.
npx @contrastlabs/cli setup
# or add the hosted MCP server to any client
# https://contrastdb.com/api/mcpaws redshift-serverless create-namespace \
--namespace-name agents
aws redshift-serverless create-workgroup \
--namespace-name agents \
--workgroup-name agents \
--base-capacity 8
# then run the MCP server on your machine
uvx awslabs.redshift-mcp-server@latestContrast and Redshift, feature by feature
| Contrast | Redshift | |
|---|---|---|
Storage and openness | ||
| Open table format by default | ContrastYes. Every synced table is Iceberg. | RedshiftPartly. Native tables use Redshift Managed Storage. |
| Tables in buckets you own | ContrastYes. On Enterprise, readable by other Iceberg tools. | RedshiftPartly. Iceberg tables in S3, cataloged in Glue. |
| Accelerate Iceberg tables you already have | ContrastYes. Connect them; they stay the source of truth. | RedshiftPartly. Queried through an external schema. |
Compute and pricing | ||
| Billed by data scanned | ContrastYes. $10 per TB after 50 GB free. | RedshiftPartly. Only Spectrum; the rest bills by time. |
| Nothing to size before querying | ContrastYes. No cluster or workgroup to pick. | RedshiftPartly. You pick nodes or a base RPU capacity. |
| Spend limits on by default | ContrastYes. Daily limits plus a monthly budget. | RedshiftPartly. Maximum RPU-hours, if you set it. |
Engine and ingestion | ||
| Query engine built for Iceberg | ContrastYes. Lakeserve, with a warm NVMe cache. | RedshiftPartly. Iceberg reads use the Redshift engine. |
| Postgres changes streamed into Iceberg | ContrastYes. Embrasure Flow streams every insert, update and delete. | RedshiftPartly. Zero-ETL replicates into Redshift tables. |
Agents | ||
| Hosted MCP server | ContrastYes. OAuth at contrastdb.com/api/mcp. | RedshiftPartly. Open source, run with your AWS credentials. |
| Agents can set up sources and ingestion | ContrastYes. MCP and CLI tools do both. | RedshiftPartly. Discovery and SQL only. |
| Read-only agent SQL by design | ContrastYes. No tool runs write SQL. | RedshiftPartly. Read-only by default; writes are opt-in. |
| Catalog context for agents | ContrastYes. Definitions, table notes and 30 days of usage. | RedshiftPartly. Databases, schemas, tables and columns. |
Running alongside Redshift
- Connect Redshift with a read-only database user over SSL, using your cluster or Serverless workgroup endpoint.
- Contrast catalogs the schemas you pick and queries them where they live, without syncing.
- Sync Postgres, Supabase and apps such as Stripe and Salesforce into Iceberg in the Contrast Warehouse.
- Give agents the hosted MCP server or the contrast CLI. Both search one catalog with your synced tables and your Redshift tables.
More comparisons
Redshift details are from its public documentation, checked November 2026: Amazon Redshift pricing, Billing for Redshift Serverless, Using Apache Iceberg tables with Redshift, Amazon Redshift and PostgreSQL, Zero-ETL integrations, Amazon Redshift MCP server, create-workgroup CLI reference. Product names are trademarks of their owners.

