.. _clickhouse-cluster: How To Run Aspects With ClickHouse Cluster ****************************************** ClickHouse clusters are an advanced way of running ClickHouse, but offer many benefits for large scale installations. Deciding whether to run in a clustered environment is a big decision with many cost and administrative impacts and should be carefully considered before launching a production environment. In most cases we expect Aspects to perform well without cluster scaling. .. warning:: Please fully understand the nuances of ClickHouse replication before choosing this route! While we have done some testing to ensure basic functionality of Aspects in a clustered environment, it may be expected to fail in novel and spectacular ways until we have a fully tested release running under load. It is strongly suggested that if you are considering clustering you use a service provider to manage the cluster for you. `ClickHouse Cloud `_ and `Altinity `_ both offer ClickHouse as a service. Altinity also offers `management services `_ for running ClickHouse in your own Kubernetes environment. Configuring Aspects to run in clustered mode is simple: - Follow the directions in :ref:`remote-clickhouse` to set up your ClickHouse connection - Set the ``CLICKHOUSE_CLUSTER_NAME`` Tutor configuration variables as appropriate for your installation - Save your Tutor configuration: ``tutor config save`` When you initialize your environment the following changes will happen from a single server install: - All users, databases, tables, views, functions, etc. will be created with ``ON CLUSTER 'your-cluster-name'``, including dbt managed views. - All tables will be created with the Replicated versions of their table engines (ex: ``ReplicatedReplacingMergeTree``) - ClickHouse traffic will be split among your nodes, as described in `their documentation `_ .