#	exclusive	inclusive	rows x	rows	loops	node
1.	630.995	1303.983	↑ 1.1	1381	1	→ HashAggregate (cost=43134.82..43149.84 rows=1502 width=12) (actual time=1303.735..1303.983 rows=1381 loops=1)
2.	209.115	672.988	↑ 1.0	836971	1	→ Hash Join (cost=139.13..38949.97 rows=836971 width=12) (actual time=5.243..672.988 rows=836971 loops=1) Hash Cond: (lpc.corpus_id_ = c.id_)
3.	324.420	460.349	↑ 1.0	836971	1	→ Hash Join (cost=45.41..26301.68 rows=836971 width=20) (actual time=1.694..460.349 rows=836971 loops=1) Hash Cond: (lpc.pair_id_ = lp.id_)
4.	134.276	134.276	↑ 1.0	836971	1	→ Seq Scan on letterpairs_corpuses lpc (cost=0.00..13701.71 rows=836971 width=24) (actual time=0.011..134.276 rows=836971 loops=1)
5.	0.942	1.653	↑ 1.0	1618	1	→ Hash (cost=25.18..25.18 rows=1618 width=12) (actual time=1.653..1.653 rows=1618 loops=1) Buckets: 1024 Batches: 1 Memory Usage: 76kB
6.	0.711	0.711	↑ 1.0	1618	1	→ Seq Scan on letterpairs lp (cost=0.00..25.18 rows=1618 width=12) (actual time=0.009..0.711 rows=1618 loops=1)
7.	1.246	3.524	↑ 1.0	2429	1	→ Hash (cost=63.36..63.36 rows=2429 width=8) (actual time=3.524..3.524 rows=2429 loops=1) Buckets: 1024 Batches: 1 Memory Usage: 95kB
8.	2.278	2.278	↑ 1.0	2429	1	→ Seq Scan on corpuses c (cost=0.00..63.36 rows=2429 width=8) (actual time=0.010..2.278 rows=2429 loops=1) Filter: (name_ = 'main'::text)

"HashAggregate  (cost=43134.82..43149.84 rows=1502 width=12) (actual time=1303.735..1303.983 rows=1381 loops=1)"
"  ->  Hash Join  (cost=139.13..38949.97 rows=836971 width=12) (actual time=5.243..672.988 rows=836971 loops=1)"
"        Hash Cond: (lpc.corpus_id_ = c.id_)"
"        ->  Hash Join  (cost=45.41..26301.68 rows=836971 width=20) (actual time=1.694..460.349 rows=836971 loops=1)"
"              Hash Cond: (lpc.pair_id_ = lp.id_)"
"              ->  Seq Scan on letterpairs_corpuses lpc  (cost=0.00..13701.71 rows=836971 width=24) (actual time=0.011..134.276 rows=836971 loops=1)"
"              ->  Hash  (cost=25.18..25.18 rows=1618 width=12) (actual time=1.653..1.653 rows=1618 loops=1)"
"                    Buckets: 1024  Batches: 1  Memory Usage: 76kB"
"                    ->  Seq Scan on letterpairs lp  (cost=0.00..25.18 rows=1618 width=12) (actual time=0.009..0.711 rows=1618 loops=1)"
"        ->  Hash  (cost=63.36..63.36 rows=2429 width=8) (actual time=3.524..3.524 rows=2429 loops=1)"
"              Buckets: 1024  Batches: 1  Memory Usage: 95kB"
"              ->  Seq Scan on corpuses c  (cost=0.00..63.36 rows=2429 width=8) (actual time=0.010..2.278 rows=2429 loops=1)"
"                    Filter: (name_ = 'main'::text)"
"Total runtime: 1304.276 ms"

Per node type stats

node type	count	sum of times	% of query
Hash	2	2.188 ms	0.2 %
Hash Join	2	533.535 ms	40.9 %
HashAggregate	1	630.995 ms	48.4 %
Seq Scan	3	137.265 ms	10.5 %

Per table stats

Table name	Scan count	Total time	% of query
scan type	count	sum of times	% of table
corpuses	1	2.278 ms	0.2 %
Seq Scan	1	2.278 ms	100.0 %
letterpairs	1	0.711 ms	0.1 %
Seq Scan	1	0.711 ms	100.0 %
letterpairs_corpuses	1	134.276 ms	10.3 %
Seq Scan	1	134.276 ms	100.0 %

explain.depesz.com

A tool for finding a real cause for slow queries.

Result: A0n

Colorize:

Columns visibility:

Per node type stats

Per table stats