---
title: Shared vs dedicated infrastructure
meta:
    description: Understand the main differences between shared and dedicated Tinybird infrastructure.
---

# Shared vs dedicated infrastructure

Tinybird runs Workspaces on either shared infrastructure or dedicated infrastructure. Both give you managed ClickHouse®, ingestion, APIs, and Tinybird tooling. The difference is how much infrastructure isolation and operational control you need.

This maps to the familiar multi-tenant versus single-tenant trade-off: shared infrastructure is multi-tenant, while dedicated infrastructure gives your Organization isolated compute.

## Shared infrastructure

On shared infrastructure, your Workspace runs on infrastructure shared with other Tinybird customers. Tinybird manages the cluster topology, ClickHouse® version, upgrades, and operational limits.

Shared infrastructure is a good fit when you want a managed experience with less operational overhead:

- You co-live on shared clusters managed by Tinybird.
- You resize vertically by choosing a larger plan size, up to the limits of shared infrastructure.
- You don't manage replicas, traffic weights, or cluster topology.
- Tinybird manages ClickHouse® versions and upgrade timing.
- Limits and safeguards protect shared resources from noisy-neighbor effects.

For billing details, see [Shared infrastructure pricing](/forward/pricing/shared-infrastructure).

## Dedicated infrastructure

On dedicated infrastructure, your organization gets an isolated ClickHouse® cluster. Your ingestion and query workloads run on replicas dedicated to you.

Dedicated infrastructure is a good fit when you need more control, isolation, or scale:

- Your workload runs on dedicated replicas, not shared compute.
- You can add, remove, resize, and rebalance replicas.
- You can control how read, write, and copy workloads are distributed across replicas.
- You have more control over ClickHouse® versioning and when upgrades happen.
- You can tune topology and capacity for workload isolation, high throughput, or stricter production requirements.

For billing details, see [Dedicated infrastructure pricing](/forward/pricing/dedicated-infrastructure). For operational controls, see [Cluster management](/forward/guides/cluster-management).

## Which one should you use?

Start with shared infrastructure if you want the simplest managed path and your workload fits within shared plan limits. Move to dedicated infrastructure when you need stronger isolation, higher scale, custom replica topology, or more control over operational changes like upgrades.