| Type: | integer |
| Default: | 1024 (8MB) |
| Min: | 0 (0kB) |
| Max: | 715827882 (5726623056kB) |
| Unit: | 8kB |
| Context: | user |
| Restart: | false |
| Since: | 10 |
Sets the minimum amount of table data that must be scanned in order for a parallel scan to be considered. For a parallel sequential scan, the amount of table data scanned is always equal to the size of the table, but when indexes are used the amount of table data scanned will normally be less. If this value is specified without units, it is taken as blocks, that is BLCKSZ bytes, typically 8kB. The default is 8 megabytes (8MB).
Recommendations
… , unless doing IoT or a read-only database. Raise to 100MB or so if your traffic on the database is very bursty, to prevent the WAL from shrinking too much.
On StackOverflow
- Nested Loop Left Join cost too much time?
- Osm2pgsql extremely slow on import on server with 192GB RAM
- Postgresql going mad
- Geoserver WFS + PostgreSQL with large table impossibly slow
- Why does "explain analyzing" this query fnish fast (showing a parallel plan), while running the actual query without "explain analyze" doesn't?
On pgsql-hackers
- [sqlsmith] parallel worker errors "subplan ... was not initialized"
- Re: pgstat_report_activity() and parallel CREATE INDEX (was: Parallel index creation & pg_stat_activity)
- RE: Parallel INSERT (INTO ... SELECT ...)
- Reigning in ExecParallelHashRepartitionFirst
- Re: Fix for parallel BTree initialization bug