- Newest
- Most votes
- Most comments
When you create your first OpenSearch Serverless collection, AWS automatically instantiates a minimum of four OCUs - two for indexing and two for search. These OCUs exist even when there's no indexing or search activity. All subsequent collections can share these OCUs (unless they use unique AWS KMS keys, which would require their own set of four OCUs).
The system is designed to scale down to a minimum of 2 OCUs for search and 2 for indexing when traffic decreases, but it cannot go below this baseline. This explains why you're seeing search capacity usage even with zero search queries.
When you try to set a maximum search capacity below the current usage, the system prevents this because it cannot go below what it's currently using (which is the minimum required for the service to function).
For all collection types, the minimum OCU capacity allowed for an account is 1 OCU (which is actually 0.5 OCU × 2) for indexing and 1 OCU (0.5 OCU × 2) for search. However, even with this minimum configuration, you'll still be charged for these baseline OCUs as they're required for the service to operate.
If this minimum cost structure doesn't align with your budget requirements, you may need to evaluate alternative solutions as you mentioned.
Sources
OpenSearch Serverless OCU usage too high | AWS re:Post
Managing capacity limits for Amazon OpenSearch Serverless - Amazon OpenSearch Service
How to enable OpenSearch Serverless 1 OCU deployment option? | AWS re:Post
Relevant content
- asked 3 years ago

This makes no sense for a serverless service. We were okay with 2 OCUs, but 10 with no search queries at all is madness, and definitely not a real serverless service.