Thread: index usage for min() vs. "order by asc limit 1"
I have two queries in PG 9.1. One uses an index like I would like, the other does not. Is this expected behavior? If
so,is there any way around it?
postgres=# explain analyze select min(id) from delayed_jobs where strand='sis_batch:account:15' group by strand;
QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------
GroupAggregate (cost=0.00..8918.59 rows=66 width=29) (actual time=226.759..226.760 rows=1 loops=1)
-> Seq Scan on delayed_jobs (cost=0.00..8553.30 rows=72927 width=29) (actual time=0.014..169.941 rows=72268
loops=1)
Filter: ((strand)::text = 'sis_batch:account:15'::text)
Total runtime: 226.817 ms
(4 rows)
postgres=# explain analyze select id from delayed_jobs where strand='sis_batch:account:15' order by id limit 1;
QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------
Limit (cost=0.00..0.33 rows=1 width=8) (actual time=0.097..0.098 rows=1 loops=1)
-> Index Scan using index_delayed_jobs_on_strand on delayed_jobs (cost=0.00..24181.74 rows=72927 width=8) (actual
time=0.095..0.095rows=1 loops=1)
Index Cond: ((strand)::text = 'sis_batch:account:15'::text)
Total runtime: 0.129 ms
(4 rows)
On Nov 17, 2011, at 5:12 PM, Ben Chobot wrote: > I have two queries in PG 9.1. One uses an index like I would like, the other does not. Is this expected behavior? If so,is there any way around it? I don't think you want the group by in that first query. Cheers, Steve > > > postgres=# explain analyze select min(id) from delayed_jobs where strand='sis_batch:account:15' group by strand; > QUERY PLAN > -------------------------------------------------------------------------------------------------------------------------- > GroupAggregate (cost=0.00..8918.59 rows=66 width=29) (actual time=226.759..226.760 rows=1 loops=1) > -> Seq Scan on delayed_jobs (cost=0.00..8553.30 rows=72927 width=29) (actual time=0.014..169.941 rows=72268 loops=1) > Filter: ((strand)::text = 'sis_batch:account:15'::text) > Total runtime: 226.817 ms > (4 rows) > > postgres=# explain analyze select id from delayed_jobs where strand='sis_batch:account:15' order by id limit 1; > QUERY PLAN > --------------------------------------------------------------------------------------------------------------------------------------------------------- > Limit (cost=0.00..0.33 rows=1 width=8) (actual time=0.097..0.098 rows=1 loops=1) > -> Index Scan using index_delayed_jobs_on_strand on delayed_jobs (cost=0.00..24181.74 rows=72927 width=8) (actual time=0.095..0.095rows=1 loops=1) > Index Cond: ((strand)::text = 'sis_batch:account:15'::text) > Total runtime: 0.129 ms > (4 rows) > > > -- > Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgsql-performance
On Nov 17, 2011, at 5:20 PM, Steve Atkins wrote:
I don't think you want the group by in that first query.
Heh, I tried to simply the example, but in reality that = becomes an in clause of multiple values. So the group by is needed.
postgres=# explain analyze select min(id) from delayed_jobs where strand='sis_batch:account:15' group by strand;QUERY PLAN--------------------------------------------------------------------------------------------------------------------------GroupAggregate (cost=0.00..8918.59 rows=66 width=29) (actual time=226.759..226.760 rows=1 loops=1)-> Seq Scan on delayed_jobs (cost=0.00..8553.30 rows=72927 width=29) (actual time=0.014..169.941 rows=72268 loops=1)Filter: ((strand)::text = 'sis_batch:account:15'::text)Total runtime: 226.817 ms(4 rows)postgres=# explain analyze select id from delayed_jobs where strand='sis_batch:account:15' order by id limit 1;QUERY PLAN---------------------------------------------------------------------------------------------------------------------------------------------------------Limit (cost=0.00..0.33 rows=1 width=8) (actual time=0.097..0.098 rows=1 loops=1)-> Index Scan using index_delayed_jobs_on_strand on delayed_jobs (cost=0.00..24181.74 rows=72927 width=8) (actual time=0.095..0.095 rows=1 loops=1)Index Cond: ((strand)::text = 'sis_batch:account:15'::text)Total runtime: 0.129 ms(4 rows)--Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)To make changes to your subscription:http://www.postgresql.org/mailpref/pgsql-performance
--
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance
can you run an analyze command first and then post here the results of: select * FROM pg_stats WHERE tablename = 'delayed_jobs'; ? -- View this message in context: http://postgresql.1045698.n5.nabble.com/index-usage-for-min-vs-order-by-asc-limit-1-tp5002928p5004410.html Sent from the PostgreSQL - performance mailing list archive at Nabble.com.