PS-11264: Vector index support in Data Dictionary#6000
Open
percona-mhansson wants to merge 3 commits into
Open
PS-11264: Vector index support in Data Dictionary#6000percona-mhansson wants to merge 3 commits into
percona-mhansson wants to merge 3 commits into
Conversation
percona-mhansson
force-pushed
the
vector-mvp-dd
branch
2 times, most recently
from
June 10, 2026 14:24
af4cd72 to
fee791a
Compare
percona-mhansson
force-pushed
the
vector-mvp-dd
branch
from
June 23, 2026 13:04
fee791a to
89e8d21
Compare
https://perconadev.atlassian.net/browse/PS-11297 validate_rec_offset() in the rec_offs_*() helpers of rem0wrec was called unconditionally, including in non-debug builds where it does no useful work. Wrap these calls in ut_d() so they are compiled only in debug builds, removing the per-call overhead on release builds.
…ad the access_time https://perconadev.atlassian.net/browse/PS-11120 This patch reduces contention inside buf_page_optimistic_get. We acquire the BUF_BLOCK_MUTEX twice there. The second time we acquire it only to update the access time, and only if it was zero. By reading the access time the first time we hold the mutex, we can skip re-acquiring it when the page has already been accessed (non-zero access time), which is the common case for hot pages. This is safe because: the second time the mutex is taken the access time is re-checked anyway; access time is never reset to zero (except when a page is reinitialized); and the page cannot be reinitialized meanwhile because it was buffer-fixed on first access. This is a contribution from Anna Glasgall <aglasgal@akamai.com> / Akamai, with a minor fix added.
Vector indexes have the type (algorithm) SE_SPECIFIC, and we add a column option in the data dictionary saying `vector_index=1;` which gets picked up by dedicated code in the data dictionary and the handler part of InnoDB. In the SQL layer, the vector index is very much a thing; there is an `HA_KEY_ALG_VECTOR`, an `HA_VECTOR` and a `KEYTYPE_VECTOR`. Extra SQL is added to display the type of a vector index as VECTOR rather than SE_SPECIFIC.
percona-mhansson
force-pushed
the
vector-mvp-dd
branch
from
July 20, 2026 09:03
89e8d21 to
cd4a559
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Vector indexes have the type (algorithm) SE_SPECIFIC, and we add a column option in the data dictionary saying
vector_index=1;which gets picked up by dedicated code in the data dictionary and the handler part of InnoDB.In the SQL layer, the vector index is very much a thing; there is an
HA_KEY_ALG_VECTOR, anHA_VECTORand aKEYTYPE_VECTOR.Extra SQL is added to display the type of a vector index as VECTOR rather than SE_SPECIFIC.
I you try to open a table containing a vector index in a trunk Percona server, you get a failed assertion in InnoDB when a client connects. However, if you drop the index, you can connect just fine.