Open RaduBerinde opened 7 years ago
Other database systems do this. We should too. For example, https://www.postgresql.org/docs/9.5/static/storage-toast.html.
@jordanlewis in which project should this land do you think?
Quoting @petermattis
While we could implement support for large values entirely via column families, I think it would be better to plumb that knowledge down into the Core layer.
We might want to consider this again in the not-so-distant-future to help people who are trying to store large values in rows, and having memory usage and other issues from doing so.
We have marked this issue as stale because it has been inactive for 18 months. If this issue is still relevant, removing the stale label or adding a comment will keep it active. Otherwise, we'll close it in 10 days to keep the issue queue tidy. Thank you for your contribution to CockroachDB!
I think there is some value to think about what it would take to have adaptive column families. Specifically, allow large values to be "pulled out" of a family to a separate KV upon insertion, on a row-by-row basis. If it helps, we could leave a special marker inside the family which indicates that the value was separated out.
Updates will get trickier, but this would allow us to have large column families by default while still avoiding very large KVs (and thus slow updates).
CC @danhhz @petermattis
Jira issue: CRDB-6118