#3549 closed enhancement (fixed)
Query Parallelization support for PostgreSQL 9.6+
| Reported by: | robe | Owned by: | pramsey |
|---|---|---|---|
| Priority: | blocker | Milestone: | PostGIS 2.3.0 |
| Component: | postgis | Version: | master |
| Keywords: | parallel, postgresql 9.6 | Cc: |
Description
pramsey this is just a reminder. Get your patch in so we can start testing against 9.6 head and I can start crash testing without having to deal with two repos.
Change History (8)
comment:1 by , 8 years ago
| Summary: | Querty Parallelization support for PostgreSQL 9.6+ → Query Parallelization support for PostgreSQL 9.6+ |
|---|
comment:2 by , 8 years ago
comment:3 by , 8 years ago
pramsey this is a threat - if you don't have this committed by the time I'm reasdy to start committing #3556 work, I will have to do this for you.
comment:4 by , 8 years ago
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
Done at r14905, keep up the good work :)
comment:5 by , 8 years ago
I'm sure there is a good reason for this. Why do the ST_Dump* ones not qualify for PARALLEL SAFE? Is it because they create rows?
comment:6 by , 8 years ago
Why are none of the
-- OGC defined marked as Parallel safe. Most look like vanilla scalar functions
e.g. ST_Dimension
comment:7 by , 8 years ago
No good reason in all cases. The only things that aren't really parallel safe are the ones w/ side effects, that write tables and stuff.
comment:8 by , 8 years ago
I committed more at r14912 , but stupid me reference the wrong ticket number and forgot to add the word references so didn't get commented anywhere.
If it's any encouragement PostgreSQL 9.6 beta1 is coming out tomorrow (or rather today) I think. It's already been tagged.