#	exclusive	inclusive	rows x	rows	loops	node
1.	34.882	51,485.899	↑ 1.0	1	1	→ Aggregate (cost=748,867,356.99..748,867,357.00 rows=1 width=8) (actual time=51,485.899..51,485.899 rows=1 loops=1)
2.	81.577	51,451.017	↓ 3.0	300,000	1	→ Nested Loop (cost=7,488.63..748,867,106.99 rows=100,000 width=0) (actual time=0.489..51,451.017 rows=300,000 loops=1)
3.	43.586	69.440	↓ 150.0	150,000	1	→ Result (cost=0.00..25.02 rows=1,000 width=32) (actual time=0.008..69.440 rows=150,000 loops=1)
4.	25.853	25.854	↓ 150.0	150,000	1	→ ProjectSet (cost=0.00..5.02 rows=1,000 width=4) (actual time=0.003..25.854 rows=150,000 loops=1)
5.	0.001	0.001	↑ 1.0	1	1	→ Result (cost=0.00..0.01 rows=1 width=0) (actual time=0.001..0.001 rows=1 loops=1)
6.	0.000	51,300.000	↑ 50.0	2	150,000	→ Nested Loop (cost=7,488.63..748,866.07 rows=100 width=0) (actual time=0.183..0.342 rows=2 loops=150,000)
7.	0.000	0.000	↑ 50.0	2	150,000	→ Function Scan on unnest key (cost=0.00..1.00 rows=100 width=32) (actual time=0.000..0.000 rows=2 loops=150,000)
8.	300.000	51,300.000	↑ 1.0	1	300,000	→ Limit (cost=7,488.63..7,488.63 rows=1 width=176) (actual time=0.170..0.171 rows=1 loops=300,000)
9.	4,800.000	51,000.000	↑ 507.0	1	300,000	→ Sort (cost=7,488.63..7,489.90 rows=507 width=176) (actual time=0.170..0.170 rows=1 loops=300,000) Sort Key: es.created_at DESC Sort Method: top-N heapsort Memory: 25kB
10.	46,200.000	46,200.000	↑ 5.5	92	300,000	→ Index Scan using foobars2 on event_series es (cost=0.56..7,486.09 rows=507 width=176) (actual time=0.008..0.154 rows=92 loops=300,000) Index Cond: (entity_id = ((((generate_series(1, 150000)))::text \|\| '_1'::text))) Filter: ((created_at < '2019-01-01 00:00:00'::timestamp without time zone) AND (key = key.key)) Rows Removed by Filter: 258

Planning time	:	0.317 ms
Execution time	:	51,485.962 ms

Aggregate  (cost=748867356.99..748867357.00 rows=1 width=8) (actual time=51485.899..51485.899 rows=1 loops=1)
  ->  Nested Loop  (cost=7488.63..748867106.99 rows=100000 width=0) (actual time=0.489..51451.017 rows=300000 loops=1)
        ->  Result  (cost=0.00..25.02 rows=1000 width=32) (actual time=0.008..69.440 rows=150000 loops=1)
              ->  ProjectSet  (cost=0.00..5.02 rows=1000 width=4) (actual time=0.003..25.854 rows=150000 loops=1)
                    ->  Result  (cost=0.00..0.01 rows=1 width=0) (actual time=0.001..0.001 rows=1 loops=1)
        ->  Nested Loop  (cost=7488.63..748866.07 rows=100 width=0) (actual time=0.183..0.342 rows=2 loops=150000)
              ->  Function Scan on unnest key  (cost=0.00..1.00 rows=100 width=32) (actual time=0.000..0.000 rows=2 loops=150000)
              ->  Limit  (cost=7488.63..7488.63 rows=1 width=176) (actual time=0.170..0.171 rows=1 loops=300000)
                    ->  Sort  (cost=7488.63..7489.90 rows=507 width=176) (actual time=0.170..0.170 rows=1 loops=300000)
                          Sort Key: es.created_at DESC
                          Sort Method: top-N heapsort  Memory: 25kB
                          ->  Index Scan using foobars2 on event_series es  (cost=0.56..7486.09 rows=507 width=176) (actual time=0.008..0.154 rows=92 loops=300000)
                                Index Cond: (entity_id = ((((generate_series(1, 150000)))::text || '_1'::text)))
                                Filter: ((created_at < '2019-01-01 00:00:00'::timestamp without time zone) AND (key = key.key))
                                Rows Removed by Filter: 258
Planning Time: 0.317 ms
Execution Time: 51485.962 ms

Per node type stats

node type	count	sum of times	% of query
Aggregate	1	34.882 ms	0.1 %
Function Scan	1	0.000 ms	0.0 %
Index Scan	1	46,200.000 ms	89.7 %
Limit	1	300.000 ms	0.6 %
Nested Loop	2	81.577 ms	0.2 %
ProjectSet	1	25.853 ms	0.1 %
Result	2	43.587 ms	0.1 %
Sort	1	4,800.000 ms	9.3 %

Per table stats

Table name	Scan count	Total time	% of query
scan type	count	sum of times	% of table
event_series	1	46,200.000 ms	89.7 %
Index Scan	1	46,200.000 ms	100.0 %

explain.depesz.com

PostgreSQL's explain analyze made readable

Result: otH6 : Optimization for: plan #lVOd

Color mode:

Visible columns:

Optimization path:

Optimization(s) for this plan:

Per node type stats

Per table stats