aws-samples / aws-samples/bedrock-chat
[Feature Request] Is there any stats available for Amazon Bedrock Knowledge Database cost?
- Dominant language
- TypeScript
- Stars
- 1.3k
- Forks
- 535
- Avg merge
- 1d 12h
- Merged PRs (30d)
- 10
Description
## Describe the solution you'd like
I set up the full project stack keeping all settings default. Among them, enableRagReplicas=true. As a consequence, all my OpenSearch collections are set up with "Deployment type = Enabled redundancy" Currently, I have two collections in place, one is 2.9mb in size (1 document), the other one 2.3mb (3 documents).
I am interested in a more cost-effective vectorstore solution, as I want to scale to thousands of documents.
## Why the solution needed
With my current setup, I have Opensearch cost of already **11.50 US$ in Region Frankfurt**. Either I have done something wrong or otherwise I would consider this cost level as not acceptable.
## Additional context
To be able to compare, I created a new Opensearch collection today, with enableRagReplicas=false. I would assume that this collection will have half of the cost than the redundant option, but I will see in a couple of days, or otherwise any comment on this cost level is appreciated. The Opensearch cost related documentation I found so far is not really going into that detail.
## Implementation feasibility
Are you willing to collaborate with us to discuss the solution, decide on the approach, and assist with the implementation?
- [ ] Yes, I am able to implement the feature and create a pull request.
- [X] No, I am unable to implement the feature, but I am open to discussing the solution.
Contributor guide
Assessment
This issue has not been assessed yet.