explain.depesz.com

A tool for finding a real cause for slow queries.

Result: bPF

options
# exclusive inclusive rows x rows loops node
1. 0.016 14.578 ↑ 1.9 30 1
→

Nested Loop (cost=10414.15..10576.52 rows=58 width=4) (actual time=14.508..14.578 rows=30 loops=1)

2. 0.011 14.502 ↑ 1.0 15 1
→

HashAggregate (cost=10414.15..10414.30 rows=15 width=4) (actual time=14.501..14.502 rows=15 loops=1)

3. 0.002 14.491 ↑ 1.0 15 1
→

Limit (cost=10413.93..10413.96 rows=15 width=12) (actual time=14.490..14.491 rows=15 loops=1)

4. 0.609 14.489 ↑ 113.8 15 1
→

Sort (cost=10413.93..10418.19 rows=1707 width=12) (actual time=14.489..14.489 rows=15 loops=1)

  • Sort Key: replays_game.game_time
  • Sort Method: top-N heapsort Memory: 25kB
5. 1.335 13.880 ↑ 1.3 1289 1
→

Nested Loop (cost=7914.79..10372.05 rows=1707 width=12) (actual time=7.040..13.880 rows=1289 loops=1)

6. 7.389 7.389 ↓ 2.9 1289 1
→

HashAggregate (cost=7914.79..7919.18 rows=439 width=4) (actual time=7.028..7.389 rows=1289 loops=1)

  • -> Index Scan using replays_playeringame_player_id on replays_playeringame (cost=0.00..7910.52 rows=1707 width=4) (actual time=0.052..5.790 rows=1289 l\
  • Index Cond: (player_id = 50027)
  • Filter: ((win IS NOT NULL) AND (apm IS NOT NULL) AND (wpm IS NOT NULL) AND (apm > 0::double precision) AND (wpm > 0::double precision))
7. 5.156 5.156 ↑ 1.0 1 1289
→

Index Scan using replays_game_pkey on replays_game (cost=0.00..5.57 rows=1 width=12) (actual time=0.004..0.004 rows=1 loops=1289)

  • Index Cond: (id = public.replays_playeringame.game_id)
8. 0.060 0.060 ↑ 2.0 2 15
→

Index Scan using replays_playeringame_game_id on replays_playeringame (cost=0.00..10.76 rows=4 width=4) (actual time=0.003..0.004 rows=2 loops=15)

  • Index Cond: (game_id = replays_game.id)