Set up docs MCPThis section describes how to manage ClickStack.
Admin guides
| Section | Description |
|---|---|
| Basic administration | An introduction to performing common administrative tasks in ClickStack. |
| Configuration options | Configuration options available for each component of ClickStack. |
| Estimating resources | Guidance on sizing compute resources for Managed ClickStack based on ingest volume. |
| Going to production | Recommended steps and best practices before running ClickStack in production. |
| Materialized views | A detailed guide to using materialized views in ClickStack to accelerate query performance. |
| Performance tuning | A comprehensive guide to tuning ClickStack for large-scale workloads. |
| RBAC | Configure role-based access control to manage team permissions for dashboards, saved searches, and sources. |
| Managing TTL | Managing data retention with TTL (time-to-live) in ClickStack. |
Core ClickHouse concepts
Most ClickStack management tasks require familiarity with the underlying ClickHouse database. We recommend reviewing the following core ClickHouse concepts outlined below before performing administrative or performance-related operations.
| Concept | Description |
|---|---|
| Tables | How ClickStack data sources map to underlying ClickHouse tables. ClickHouse tables primarily use the MergeTree engine. |
| Parts | How data is written as immutable parts and merged over time. |
| Partitions | Logical groupings of table parts that simplify data management, querying, and optimization. |
| Merges | The background process that combines parts to reduce the number of parts queried and maintain performance. |
| Granules | The smallest unit of data read and pruned during query execution. |
| Primary (ordering) keys | How the ORDER BY key defines on-disk data layout, compression, and query pruning behavior. |
These concepts are fundamental to ClickHouse performance and will help you make informed administrative decisions when managing ClickStack.