Re: [GENERAL] Perfomance of IN-clause with many elements and possible solutions

regression=# EXPLAIN (ANALYZE, BUFFERS) SELECT count(*) FROM ids WHERE id IN
:values_clause;
                                                            QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------
 Aggregate  (cost=245006.16..245006.17 rows=1 width=8) (actual time=3550.581..3550.581 rows=1 loops=1)
 Execution time: 3550.700 ms

regression=# set enable_hashagg TO 0;
regression=# set enable_sort TO 0;
SET
regression=# EXPLAIN (ANALYZE, BUFFERS) SELECT count(*) FROM ids WHERE id IN
:values_clause;
                                                          QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------
 Aggregate  (cost=320003.90..320003.91 rows=1 width=8) (actual time=3548.364..3548.364 rows=1 loops=1)
 Execution time: 3548.463 ms

At least in this example, the actual runtimes are basically identical
regardless, so there is no great point in sweating over it.