Commit 8b49a60

committed

Cache catalog lookup data across groups in ordered-set aggregates.

The initial commit of ordered-set aggregates just did all the setup work afresh each time the aggregate function is started up. But in a GROUP BY query, the catalog lookups need not be repeated for each group, since the column datatypes and sort information won't change. When there are many small groups, this makes for a useful, though not huge, performance improvement. Per suggestion from Andrew Gierth. Profiling of these cases suggests that it might be profitable to avoid duplicate lookups within tuplesort startup as well; but changing the tuplesort APIs would have much broader impact, so I left that for another day.

1 parent 92459e7 commit 8b49a60Copy full SHA for 8b49a60

2 files changed

+290

-218

lines changed

src
- backend/utils/adt
  - orderedsetaggs.c
- include/catalog
  - pg_operator.h

2 files changed

+290

-218

lines changed

Comments

(0)

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Commit 8b49a60

2 files changed

2 files changed

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

File tree

2 files changed

2 files changed

0 commit comments

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.