Feedback: Database Schemas - perhaps incorrect/unhelpful guidance on large number of schemas · Issue #19624 · cockroachdb/docs · GitHub
More Web Proxy on the site http://driver.im/
You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The number of databases or schemas on the cluster has minimal impact on the total number of tables that it can support.
This implies that a CockroachDB cluster can support a large number of schemas, which is not the case. Please consult with the engineering team to reword this in a way that more accurately reflects the impact of large numbers of non-table objects, including schemas. Anecdotally, I have heard that CockroachDB has problems with a large total number of object descriptors, rather than the numbers of specific types of database objects.
Note that different subsystems may be impacted differently. So, for example, query performance may be acceptable but other database components, such as the DB Console or backups, may fail or perform poorly.
As a specific example for schemas, performed a simple test. I created 20000 schemas on a CockroachDB cluster. SQL queries against tables in these schemas seemed to perform fine, but a SHOW SCHEMAS query took an extremely long time to run.
Uh oh!
There was an error while loading. Please reload this page.
Page: https://cockroachlabs.com/docs/v25.2/schema-design-overview.html
What is the reason for your feedback?
[ ] Missing the information I need
[ ] Too complicated
[ ] Out of date
[ ] Something is broken
[X] Other
Additional details
The page says
This implies that a CockroachDB cluster can support a large number of schemas, which is not the case. Please consult with the engineering team to reword this in a way that more accurately reflects the impact of large numbers of non-table objects, including schemas. Anecdotally, I have heard that CockroachDB has problems with a large total number of object descriptors, rather than the numbers of specific types of database objects.
Note that different subsystems may be impacted differently. So, for example, query performance may be acceptable but other database components, such as the DB Console or backups, may fail or perform poorly.
As a specific example for schemas, performed a simple test. I created 20000 schemas on a CockroachDB cluster. SQL queries against tables in these schemas seemed to perform fine, but a SHOW SCHEMAS query took an extremely long time to run.
THANK YOU
Jira issue: DOC-13736
The text was updated successfully, but these errors were encountered: