Relationships between resources in Managed Service for ClickHouse

The main entity used in Managed Service for ClickHouse is a database cluster.

Each cluster consists of one or more database hosts, which are virtual machines with DBMS servers deployed on them. Cluster hosts may reside in different availability zones and even different availability regions. More about the geo scope of Yandex Cloud.

• A cluster consisting of multiple hosts has automatic failover, because the hosts can take over each other's functions as the cluster's primary replica.

  When creating such a cluster, choose one of the available mechanisms to ensure replication:

  • ClickHouse Keeper.

    The cluster will consist only of ClickHouse hosts.

  • ZooKeeper.

    The cluster will consist of ClickHouse hosts and individual ZooKeeper hosts.

• A single-host cluster doesn't guarantee fault tolerance.

  Later you can add ZooKeeper hosts to this cluster to enable fault tolerance and replication. This will allow you to add multiple ClickHouse hosts to the cluster.

ClickHouse Keeper or ZooKeeper hosts manage replication and distribution of queries across cluster hosts. For more information, see Replication ClickHouse.

When creating a cluster, specify:

• Host class: A VM template for deploying the cluster hosts. For a list of available host classes and their characteristics, see Host classes in ClickHouse.

• Environment: The environment where the cluster will be deployed:

  • PRODUCTION: For stable versions of your apps.
  • PRESTABLE: For testing, including the Managed Service for ClickHouse service itself. The Prestable environment is first updated with new features, improvements, and bug fixes. However, not every update ensures backward compatibility.

You should also specify the host class when creating a shard.

A database cluster created in a folder can be accessed by all VMs connected to the same cloud network. Learn more about networking.