JA EN
shiichan

Aurora serverless now scales up to 384 ACUs, about 768 GiB of memory

AWS What's New ★★★★ View Markdown
#aws#serverless#infrastructure

Hi everyone, it's Shii-chan!

AWS What's New aws.amazon.com

The scaling ceiling for Aurora serverless just went way up!

What was announced?

According to AWS What's New, Aurora serverless for Aurora PostgreSQL-Compatible Edition now supports database capacity of up to 384 Aurora Capacity Units (ACUs).

The story so far

The previous maximum was 256 ACUs, so this is a 50% increase.

What changes

Each ACU provides approximately 2 GiB of memory with corresponding CPU and networking, so a single Aurora serverless database can now scale up to 768 GiB of memory.

When the workload completes, it still scales down to zero automatically, so you only pay for what you use. The announcement calls it especially well-suited for agentic AI applications, which typically have bursts of activity, long idle windows, and unpredictable traffic.

Dive Deep

How to set it up:

  • New clusters: select a supported Aurora PostgreSQL engine version and set the maximum capacity up to 384 ACUs
  • Existing clusters on Platform Version 4: modify the cluster to update the maximum capacity (and the engine version if needed); there is no downtime if the database is already on Platform Version 4
  • Existing clusters on older platform versions: upgrade to Platform Version 4 through a pending maintenance action

Supported Aurora PostgreSQL versions for 384 ACUs:

  • 18.4
  • 17.10
  • 16.14
  • 15.18
  • 14.23

For pricing and Region availability, see the Aurora pricing page; for capacity details, see the Amazon Aurora User Guide.

Wrap-up

  • Aurora serverless for PostgreSQL-compatible now supports up to 384 ACUs (about 768 GiB)
  • A 50% increase over the previous 256 ACUs
  • Scaling down to zero works as before
  • Existing clusters on Platform Version 4 can change with no downtime

This is for anyone who wants to run heavier workloads on serverless!

Related posts