Multitenancy

Overview

This document describes concepts related to multitenancy in Pinecone indexes. This includes information on different approaches to keeping sets of vectors separate within a Pinecone index. To learn how to create or modify an index, see Manage indexes.

You may need to segment vectors by customer or otherwise either physically or logically. This document describes different techniques to accomplish this and the advantages and advantages of each approach.

Namespaces

One approach to multitenancy is to use namespaces to isolate segments of vectors within a single index. This is a 'pool' model that shares most resources between tenants while keeping them logically separate.

Advantages

• Namespaces reduce the need for additional indexes, reducing maintenance effort.

Disadvantages

• Customer data is not isolated in its own pod. This means tenants share compute and storage resources.
• Deleting tenant data is more complicated and takes more time.

Metadata filtering

This approach to multitenancy stores all segments of vectors in a single index and filters by metadata at query time. This is another 'pool' model; here, you separate tenants on the query level.

Advantages

• Metadata filtering allows you to query across multiple tenants.
• Metadata filtering reduces the need for additional indexes, reducing maintenance effort.

Disadvantages

• Customer data is not isolated in its own pod. This means tenants share compute and storage resources.
• There is no way to track tenant-specific costs.
• You cannot provision tenants with different dimensions or pod type characteristics, which are set at the index level.

Separate indexes

Another approach to multitenancy is to create a separate index for each segment. This is a 'silo' model that provides dedicated resources to each tenant. For example, if you need to separate vectors for each customer, you can create a separate index for each customer.

Advantages

• This model separates tenants into different pods, preventing queries across tenants and allocating compute and storage resources to each tenant.

Disadvantages

• Creating and maintaining multiple indexes requires additional effort and cost.
• Because this model doesn't share resources between tenants, you must assign each tenant at minimum one pod. This is extremely inefficient if you have tenants with small number of vectors.
• Creating a new index takes more time than creating a namespace.