How many tables can you have in DynamoDB?
2,500 per Region by default. Each AWS account starts with a quota of 2,500 DynamoDB tables per Region, which AWS can raise to a maximum of 10,000 on request. Individual tables are unlimited in size — and beyond 10,000 tables, AWS recommends spreading workloads across multiple accounts.
The table quota
The initial quota is 2,500 tables per AWS Region per account. If you need more, AWS's guidance is to contact your account team for an increase up to 10,000; past that, the recommended practice is multiple accounts, each serving up to 10,000 tables.
Tables themselves are unlimited
There is no practical limit on a single table's size — tables are unconstrained in the number of items and bytes. The hard per-item limit is 400 KB, and each table supports up to 20 global secondary indexes (default quota) and 5 local secondary indexes.
What stops you long before the quota
Almost nobody reaches 2,500 tables. What bites first is that a provisioned table is never free, even when nothing touches it. AWS sets the minimum for any table or GSI at 1 read capacity unit and 1 write capacity unit, and that is billed hourly whether or not a single request arrives.
At us-east-1 rates, $0.00013 per RCU-hour and $0.00065 per WCU-hour over 730 hours, an empty minimum-capacity table costs $0.5694 a month:
| Tables at the minimum | Monthly cost before any request |
|---|---|
| 100 | $56.94 |
| 2,500 (the default quota) | $1,423.50 |
| 10,000 (the raised limit) | $5,694.00 |
Figures from our pricing calculator, which runs on the synced AWS pricing table, and before any free-tier allowance.
On-demand tables have no such floor. An idle on-demand table bills storage only, which is what makes a high table count survivable at all. If your design genuinely wants hundreds of tables, that choice of capacity mode matters more than the quota does.
Do you actually need many tables?
In DynamoDB, table count is a design choice, not a growth axis. Single-table design co-locates related entities in one table on purpose, and most applications need far fewer tables than their relational schema had — often one per service.
Go deeper
Read single-table design before splitting entities across tables, check item sizes with the item size calculator, and download DynoTable to keep an overview of every table across your profiles and Regions.
References
- Quotas in Amazon DynamoDB — Amazon DynamoDB Developer Guide
- Core components of Amazon DynamoDB — Amazon DynamoDB Developer Guide
- Local secondary indexes — Amazon DynamoDB Developer Guide
Last verified 2026-07-13 against the official AWS documentation linked above; every quota on this page re-checked 2026-07-28.
Costs computed from the us-east-1 provisioned capacity rates in our synced AWS pricing table, at the 730-hour month the calculator uses.