Re: Slow join over three tables

Поиск
Список
Период
Сортировка
От Victor Yegorov
Тема Re: Slow join over three tables
Дата
Msg-id CAGnEboiE7WkPisKRJaQ8qKUunA9TWjEXna8pj78dp-VwZWLLJA@mail.gmail.com
обсуждение исходный текст
Ответ на Slow join over three tables  (Tim van der Linden <tim@shisaa.jp>)
Ответы Re: Slow join over three tables  (Tim van der Linden <tim@shisaa.jp>)
Список pgsql-general
2016-04-27 2:27 GMT+03:00 Tim van der Linden <tim@shisaa.jp>:
The plan:

Sort  (cost=105773.63..105774.46 rows=333 width=76) (actual time=5143.162..5143.185 rows=448 loops=1)
   Sort Key: r.created
   Sort Method: quicksort  Memory: 60kB
   ->  Nested Loop  (cost=1.31..105759.68 rows=333 width=76) (actual time=54.784..5142.872 rows=448 loops=1)
     Join Filter: (d.rid = a.rid)
     ->  Nested Loop  (cost=0.87..94657.59 rows=14005 width=72) (actual time=0.822..2038.952 rows=14199 loops=1)
         ->  Index Scan using report_drugs_drug_idx on report_drugs d  (cost=0.44..500.28 rows=14005 width=31) (actual time=0.669..3.900 rows=14199 loops=1)
               Index Cond: (drug = ANY ('{359,360,361,362,363}'::integer[]))
         ->  Index Scan using reports_id_key on reports r  (cost=0.43..6.71 rows=1 width=41) (actual time=0.143..0.143 rows=1 loops=14199)
               Index Cond: (id = d.rid)
     ->  Index Scan using report_adverses_rid_idx on report_adverses a  (cost=0.44..0.78 rows=1 width=12) (actual time=0.218..0.218 rows=0 loops=14199)
           Index Cond: (rid = r.id)
           Filter: (adverse = ANY ('{"back pain - nonspecific","nonspecific back pain","back pain"}'::text[]))
           Rows Removed by Filter: 5
Planning time: 13.994 ms
Execution time: 5143.235 ms

Can you post output of `EXPLAIN (analyze, buffers)`, please?
It'd be good to check how many buffers are hit/read during Index Scans.

I have had a somewhat similar case, when due to misconfigured autovacuum (not aggressive enough for the increased volume)
I got bloat accumulated in the indexes. Perhaps, REINDEX can help here.


--
Victor Y. Yegorov

В списке pgsql-general по дате отправления:

Предыдущее
От: Jeff Janes
Дата:
Сообщение: Re: Does this perf output seem 'normal'?
Следующее
От: Francisco Olarte
Дата:
Сообщение: Re: psql color hostname prompt