pgvector and Cloudberry Database version
The GUC ivfflat.max_probes is used to control the number of results like limit, but in cbdb, data is distributed in different segments, the GUC should be adapted in MPP env.
What happened
GUC ivfflat.max_probes in cbdb env.
What you think should happen instead
No response
How to reproduce
make installcheck
Operating System
CentOS
Anything else
No response
Are you willing to submit PR?
Code of Conduct
pgvector and Cloudberry Database version
The GUC
ivfflat.max_probesis used to control the number of results likelimit, but in cbdb, data is distributed in different segments, the GUC should be adapted in MPP env.What happened
GUC
ivfflat.max_probesin cbdb env.What you think should happen instead
No response
How to reproduce
make installcheck
Operating System
CentOS
Anything else
No response
Are you willing to submit PR?
Code of Conduct