What is the best way to handle thousands of clients requiring up to 3 indexes (vector stores) per client?

Support
, , ,
1

Hello,

Now as I’m hitting the limit of 100 serverless indexes per project I need to find a solution to onboard 10000 clients (each requiring up to 3 vector stores).

What is the best approach to this? considering limitations of 100 indexes per project?

2

Hi @team2. Welcome to the Pinecone forum!

You should consider using namespaces. Namespaces let you partition records within an index. In your case, you could use a dedicated namespace for each client. You can have up to 10,000 namespaces per serverless index, so you could even handle your 10,000 clients with a single serverless index!

See our tutorial on implementing multitenancy with namespaces for more details, and please let me know if you have any questions.

Best,
Jesse