DynamoDB tables better suited to Standard-IA
When storage dominates a table's bill — lots of data, little access — the infrequent-access class costs less even though requests cost more.
What this problem is
DynamoDB offers two table classes. Standard balances storage and throughput pricing. Standard-IA charges considerably less per gigabyte stored and about 25% more per read and write unit — it is an explicit trade between holding and accessing.
For most tables Standard is the right call: throughput dominates the bill. But there is a profile where the relationship inverts — large tables with infrequent access, such as history, audit trails, archived events or rarely-queried reference data. There storage is the bulk of the bill and Standard-IA comes out cheaper even with pricier requests.
Identifying that means comparing both parts, not looking at one: it is the ratio between what the table holds and what it moves that decides. The class change is online and reversible, though AWS limits how often it can be done.
How we detect it
We list tables with ListTables and DescribeTable, which report each one's stored size and current class, and compare those against the consumed throughput recorded in CloudWatch — it is the ratio between the two that indicates which class costs less.
- DynamoDB tables better on Standard-IA
What it costs
Standard storage runs about $0.25 per GB-month and Standard-IA about $0.10 per GB-month, while Standard-IA throughput costs about 25% more (us-east-1). The crossover point depends on the ratio of storage to traffic.
Public AWS list prices, us-east-1, subject to change. The figure in your report is computed against your resources in your regions.
How it appears in the report
An illustrative example, with invented resources and figures. In the report each finding carries the real resource, its region, and the fix command where a safe one exists.
Does your account have this?
The scan is read-only and shows you what it found before you pay for anything.
Run a free scan