Free tools Windows power users keep installed

One-click scans. No signup required.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Qdrant documents tiered multitenancy from version 1.16.0. It lets smaller tenants share a fallback shard while larger tenants can be promoted to dedicated shards within the same collection. Routing targets a tenant’s dedicated shard when one is active and falls back to the shared shard otherwise.

What tiered multitenancy changes

Qdrant offers several ways to separate tenant data. With payload partitioning, tenants share a collection and queries filter on a tenant field. With user-defined sharding, a tenant can have its own shard, which offers stronger isolation but adds shard overhead. Tiered multitenancy combines the two: smaller tenants share storage, while tenants with greater needs can move to dedicated shards. Qdrant describes this as a way to handle tenants of uneven sizes; it is not an independently verified performance or cost guarantee. Qdrant’s multitenancy documentation describes the feature, and its v1.16 release article provides release context.

How Qdrant routes tenant requests

The design has three parts: user-defined shards, a shared fallback shard, and tenant promotion. Qdrant’s documentation summarizes the design this way: “With tiered multitenancy, you can implement two levels of tenant isolation within a single collection.”

Shared fallback shard

Start with a custom-sharded collection and a named fallback shard. In Qdrant’s example, the fallback shard is named default. Tenants without a dedicated shard use this shared destination.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Dedicated shards and fallback routing

Requests identify both a target shard and a fallback shard. If the target tenant shard exists and is active, Qdrant routes the request there; if not, the request goes to the fallback. Queries should use the same shard-key selector and include a filter on the tenant field. The filter value must match the target shard key.

Promoting a tenant

As a tenant grows, it can be moved from the fallback shard to a dedicated shard. Qdrant says promotion uses its internal shard-transfer mechanism and supports read and write requests. The documented pattern requires promotion to dedicated shards from a single-shard setup; Qdrant permits a replication factor greater than one.

Choose the tenancy model that fits your tenants

Model How tenant data is separated Best fit described by Qdrant Trade-off
Payload partitioning Tenants share a collection; queries filter by a tenant payload field. Many small tenants of similar size. Requires tenant filters in queries.
User-defined sharding A tenant is assigned a dedicated shard. A smaller number of larger tenants. Stronger isolation, with added shard resource overhead.
Tiered multitenancy Small tenants share a fallback shard; larger tenants can have dedicated shards in the same collection. A mix of tenant sizes, where some may grow beyond shared placement. Requires consistent routing and tenant filtering; it combines shared and dedicated placement rather than removing operational responsibilities.
Separate collections Each tenant, or group of tenants, can have its own collection. A limited tenant count when strict isolation is needed. Every collection carries resource overhead. Qdrant Cloud’s current multitenancy documentation lists a default limit of 1000 collections per cluster; this is a documented Cloud default, not a universal deployment limit.

These trade-offs and the Cloud default are described in Qdrant’s multitenancy documentation.

Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

Configuration and security considerations

  • Plan for the custom-sharded collection and its single-shard starting configuration if you expect to promote tenants to dedicated shards later.
  • Use the tenant’s shard key consistently for routing, and retain the tenant-field filter in queries. A named shard is a placement mechanism, not an authorization boundary.
  • Enforce tenant access in application logic and ensure that requests cannot substitute another tenant’s shard key or filter value.
  • Consider shard and collection overhead alongside isolation needs; neither dedicated shards nor separate collections are free of resource costs.

Qdrant’s Production & Operations documentation provides broader operational context. Tiered multitenancy is a Qdrant feature, not a requirement to use Qdrant Cloud.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.