Opened 8 years ago
Closed 5 years ago
#3562 closed task (worksforme)
Revise regress test suite to support testing of parallelization
| Reported by: | robe | Owned by: | robe |
|---|---|---|---|
| Priority: | medium | Milestone: | PostGIS 3.0.0 |
| Component: | postgis | Version: | master |
| Keywords: | Cc: |
Description
As noted in #3561 I had to modify global settings on my instance to force parallelism to kick in.
It would be really nice to be able to trigger this for some tests and also conditionally based on if pg 9.6+
Change History (8)
comment:1 by , 8 years ago
comment:2 by , 8 years ago
In theory it should be because the final worker has to merge all the results from slave workers. So if the final has an ORDER BY, the results should be ordered.
comment:3 by , 8 years ago
| Milestone: | PostGIS 2.3.0 → PostGIS 2.4.0 |
|---|
comment:4 by , 8 years ago
| Type: | defect → task |
|---|
comment:5 by , 7 years ago
| Milestone: | PostGIS 2.4.0 → PostGIS 2.5.0 |
|---|
comment:6 by , 7 years ago
| Owner: | changed from to |
|---|
comment:7 by , 6 years ago
| Milestone: | PostGIS 2.5.0 → PostGIS 3.0.0 |
|---|
comment:8 by , 5 years ago
| Resolution: | → worksforme |
|---|---|
| Status: | new → closed |
Postgres 12 ships with parallelization enabled by default.
Note:
See TracTickets
for help on using tickets.
Would a parallel run be any predictable in term of output order ?