Explanation
Why the waste happens and who it affects.
For workloads with frequent read and write activity, this can lead to unnecessarily high costs. Aurora I/O-Optimized eliminates I/O charges entirely and simplifies cost predictability. In environments with consistently high I/O usage, switching to I/O-Optimized often results in lower total spend.
Billing model
The pricing dimensions that drive this cost.
Aurora offers two storage configurations with distinct cost structures:
- Aurora Standard: Lower per-GB storage cost, but incurs additional charges for I/O operations (reads and writes)
- Aurora I/O-Optimized: Higher per-GB storage cost and higher database instance and Aurora Serverless ACU prices, but no charges for read and write I/O operations
Choosing the right configuration depends on the balance between storage volume and I/O intensity. High-I/O workloads may see significantly lower costs under I/O-Optimized pricing; AWS positions it for clusters where I/O spend exceeds 25% of total Aurora database spend.
How to detect
7 checks to find it in your estate.
- Identify Aurora clusters currently using the Standard storage configuration
- Review historical cost breakdowns to evaluate the portion of spend attributed to I/O charges
- Determine whether the workload exhibits high read/write activity or transactional behavior
- Assess whether I/O charges regularly account for a significant share of total Aurora costs
- Estimate whether switching to I/O-Optimized would lower total cost based on current storage and I/O usage patterns
- Confirm that the cluster meets version and engine requirements for I/O-Optimized eligibility
- Engage database administrators or workload owners to validate I/O behavior and review trade-offs
How to fix
How to remove the waste.
Modify the cluster to use Aurora I/O-Optimized if the workload is read/write intensive and I/O costs are high. For supported Aurora versions, the change causes no downtime on non-NVMe-based instances (including Aurora Serverless), while NVMe-based instance classes require a database engine restart that may cause brief downtime. You can switch from Standard to I/O-Optimized only once every 30 days. Monitor ongoing cost and performance to confirm the change results in expected benefits.
Documentation
Vendor references for pricing and configuration.