#	exclusive	inclusive	rows x	rows	loops	node
1.	4.958	1952.349	↑ 2.3	67	1	→ HashAggregate (cost=251315.39..251316.96 rows=157 width=4) (actual time=1952.282..1952.349 rows=67 loops=1)
2.	13.517	1947.391	↑ 29.4	3322	1	→ Nested Loop (cost=1678.36..250826.31 rows=97817 width=4) (actual time=39.496..1947.391 rows=3322 loops=1)
3.	878.785	1910.394	↑ 32.1	2935	1	→ Hash Left Join (cost=1678.36..75649.71 rows=94105 width=8) (actual time=39.477..1910.394 rows=2935 loops=1) Hash Cond: ((crowdbreaks_tweet.place_id)::text = (crowdbreaks_place.place_id)::text) Filter: ((crowdbreaks_tweet.coordinates @ '0103000020E61000000100000005000000AE47E17A141E5FC00000000000003840AE47E17A141E5FC029ED0DBE30B14840A4703D0AD7A350C029ED0DBE30B14840A4703D0AD7A350C00000000000003840AE47E17A141E5FC00000000000003840'::geometry) OR ((crowdbreaks_place.bounding_box && '0103000020E61000000100000005000000AE47E17A141E5FC00000000000003840AE47E17A141E5FC029ED0DBE30B14840A4703D0AD7A350C029ED0DBE30B14840A4703D0AD7A350C00000000000003840AE47E17A141E5FC00000000000003840'::geometry) AND _st_overlaps(crowdbreaks_place.bounding_box, '0103000020E61000000100000005000000AE47E17A141E5FC00000000000003840AE47E17A141E5FC029ED0DBE30B14840A4703D0AD7A350C029ED0DBE30B14840A4703D0AD7A350C00000000000003840AE47E17A141E5FC00000000000003840'::geometry)))
4.	992.652	992.652	↓ 1.0	801934	1	→ Index Scan using crowdbreaks_tweet_crreated_at on crowdbreaks_tweet (cost=0.00..36121.24 rows=799554 width=125) (actual time=0.019..992.652 rows=801934 loops=1) Index Cond: ((created_at > '2012-04-18 12:47:19.550875+00'::timestamp with time zone) AND (created_at < '2012-04-19 12:47:19.550882+00'::timestamp with time zone))
5.	21.998	38.957	↓ 1.0	15848	1	→ Hash (cost=521.38..521.38 rows=15838 width=469) (actual time=38.957..38.957 rows=15848 loops=1) Buckets: 1024 Batches: 8 Memory Usage: 322kB
6.	16.959	16.959	↓ 1.0	15848	1	→ Seq Scan on crowdbreaks_place (cost=0.00..521.38 rows=15838 width=469) (actual time=0.007..16.959 rows=15848 loops=1)
7.	23.480	23.480	↑ 1.0	1	2935	→ Index Scan using crowdbreaks_incomingkeyword_tweet_id on crowdbreaks_incomingkeyword (cost=0.00..1.85 rows=1 width=12) (actual time=0.007..0.008 rows=1 loops=2935) Index Cond: (tweet_id = crowdbreaks_tweet.tweet_id)

                                                                                                                                                    QUERY PLAN                                                                                                                                                                                                                                                                                                                                                                                             
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 HashAggregate  (cost=251315.39..251316.96 rows=157 width=4) (actual time=1952.282..1952.349 rows=67 loops=1)
   ->  Nested Loop  (cost=1678.36..250826.31 rows=97817 width=4) (actual time=39.496..1947.391 rows=3322 loops=1)
         ->  Hash Left Join  (cost=1678.36..75649.71 rows=94105 width=8) (actual time=39.477..1910.394 rows=2935 loops=1)
               Hash Cond: ((crowdbreaks_tweet.place_id)::text = (crowdbreaks_place.place_id)::text)
               Filter: ((crowdbreaks_tweet.coordinates @ '0103000020E61000000100000005000000AE47E17A141E5FC00000000000003840AE47E17A141E5FC029ED0DBE30B14840A4703D0AD7A350C029ED0DBE30B14840A4703D0AD7A350C00000000000003840AE47E17A141E5FC00000000000003840'::geometry) OR ((crowdbreaks_place.bounding_box && '0103000020E61000000100000005000000AE47E17A141E5FC00000000000003840AE47E17A141E5FC029ED0DBE30B14840A4703D0AD7A350C029ED0DBE30B14840A4703D0AD7A350C00000000000003840AE47E17A141E5FC00000000000003840'::geometry) AND _st_overlaps(crowdbreaks_place.bounding_box, '0103000020E61000000100000005000000AE47E17A141E5FC00000000000003840AE47E17A141E5FC029ED0DBE30B14840A4703D0AD7A350C029ED0DBE30B14840A4703D0AD7A350C00000000000003840AE47E17A141E5FC00000000000003840'::geometry)))
               ->  Index Scan using crowdbreaks_tweet_crreated_at on crowdbreaks_tweet  (cost=0.00..36121.24 rows=799554 width=125) (actual time=0.019..992.652 rows=801934 loops=1)
                     Index Cond: ((created_at > '2012-04-18 12:47:19.550875+00'::timestamp with time zone) AND (created_at < '2012-04-19 12:47:19.550882+00'::timestamp with time zone))
               ->  Hash  (cost=521.38..521.38 rows=15838 width=469) (actual time=38.957..38.957 rows=15848 loops=1)
                     Buckets: 1024  Batches: 8  Memory Usage: 322kB
                     ->  Seq Scan on crowdbreaks_place  (cost=0.00..521.38 rows=15838 width=469) (actual time=0.007..16.959 rows=15848 loops=1)
         ->  Index Scan using crowdbreaks_incomingkeyword_tweet_id on crowdbreaks_incomingkeyword  (cost=0.00..1.85 rows=1 width=12) (actual time=0.007..0.008 rows=1 loops=2935)
               Index Cond: (tweet_id = crowdbreaks_tweet.tweet_id)
 Total runtime: 1952.471 ms
(13 rows)

Per node type stats

node type	count	sum of times	% of query
Hash	1	21.998 ms	1.1 %
Hash Left Join	1	878.785 ms	45.0 %
HashAggregate	1	4.958 ms	0.3 %
Index Scan	2	1016.132 ms	52.0 %
Nested Loop	1	13.517 ms	0.7 %
Seq Scan	1	16.959 ms	0.9 %

Per table stats

Table name	Scan count	Total time	% of query
scan type	count	sum of times	% of table
crowdbreaks_incomingkeyword	1	23.480 ms	1.2 %
Index Scan	1	23.480 ms	100.0 %
crowdbreaks_place	1	16.959 ms	0.9 %
Seq Scan	1	16.959 ms	100.0 %
crowdbreaks_tweet	1	992.652 ms	50.8 %
Index Scan	1	992.652 ms	100.0 %

explain.depesz.com

A tool for finding a real cause for slow queries.

Result: meQ

Colorize:

Columns visibility:

Per node type stats

Per table stats