Amazon Redshift adds rg.large and rg.12xlarge — up to 2.4x faster, 30% cheaper!
Hi everyone, it's Shiichan! Today I've got news about an expanded instance lineup for Amazon Redshift.
AWS What's NewWhat was announced?
According to AWS's What's New, Amazon Redshift now offers two new RG instance sizes, rg.large and rg.12xlarge, generally available today. Like the existing RG instances, they're Graviton-powered and deliver up to 2.4x faster query performance than previous-generation RA3 instances at 30% lower price per vCPU.
The story so far
Redshift's RG instance family already existed, but the lineup of sizes was limited, which sometimes made it hard to find a size that fit your workload exactly. For teams planning a migration from RA3, having fewer size options made it harder to land on a configuration that felt just right.
What changes
With rg.large and rg.12xlarge added to the lineup, you can now dial in your cluster size more precisely, whether your workload is small or large. Teams on RA3 can migrate using Snapshot and Restore, Elastic Resize, or Classic Resize, making it easier to land on the best combination of performance and price.
Dive Deep
The new rg.large and rg.12xlarge sizes are available on the current track (P202) only. Customers on the trailing track (P201) will continue using rg.xlarge and rg.4xlarge.
Pricing options include On-Demand as well as 1-year and 3-year Reserved Instances with no upfront payment. The sizes are rolling out across a wide range of regions, including major ones like US East (N. Virginia), Europe (Ireland), and Asia Pacific (Tokyo).
If you're considering a migration, it's worth checking the RA3-to-RG upgrade guide and the Redshift pricing page.
Wrap-up
- Amazon Redshift adds two new instance sizes: rg.large and rg.12xlarge
- Up to 2.4x faster query performance than previous-generation RA3 instances at 30% lower price per vCPU
- Available on the current track (P202) only; trailing track (P201) continues with rg.xlarge and rg.4xlarge
- RA3 clusters can migrate via Snapshot and Restore, Elastic Resize, or Classic Resize
- Supports On-Demand plus 1-year and 3-year Reserved Instances with no upfront payment
A great update for data engineers who want to fine-tune their cluster size to match their workload!