Thread: Trigram is slow when 10m rows

I have a simple table with Trigram index,

create table mytable(hash char(40), title text);
create index title_trgm_idx on mytable using gin(title gin_trgm_ops);

When I run a query with 10m rows, it uses the Trigram index, but takes
3s to execute, very slow.
(I have 80m rows, but only inserted 10m for testing purpose)

test=# select count(*) from mytable;
  count
----------
 13971887
(1 row)

test=# explain analyze select * from mytable where title ilike 'x264';
                                                              QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------
 Bitmap Heap Scan on mytable  (cost=462.69..5639.67 rows=1380
width=83) (actual time=2937.308..2937.308 rows=0 loops=1)
   Recheck Cond: (title ~~* 'x264'::text)
   Rows Removed by Index Recheck: 11402855
   Heap Blocks: exact=39557 lossy=158010
   ->  Bitmap Index Scan on title_trgm_idx  (cost=0.00..462.35
rows=1380 width=0) (actual time=342.440..342.440 rows=1220793 loops=1)
         Index Cond: (title ~~* 'x264'::text)
 Planning time: 0.611 ms
 Execution time: 2937.729 ms
(8 rows)

Any ideas to speed things up?

--
Best Regards,
Aaron Lewis - PGP: 0x13714D33 - http://pgp.mit.edu/
Finger Print:   9F67 391B B770 8FF6 99DC  D92D 87F6 2602 1371 4D33


--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Thanks Oleg.

I've increased work_mem to 128MB, now the query falls down to 1.7s,
faster but still not good enough.

Is there any other thing I can do about it?

test=# explain analyze select * from mytable where title ilike 'x264';
                                                              QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------
 Bitmap Heap Scan on mytable  (cost=462.69..5639.67 rows=1380
width=83) (actual time=1754.656..1754.656 rows=0 loops=1)
   Recheck Cond: (title ~~* 'x264'::text)
   Rows Removed by Index Recheck: 1220793
   Heap Blocks: exact=197567
   ->  Bitmap Index Scan on title_trgm_idx  (cost=0.00..462.35
rows=1380 width=0) (actual time=346.663..346.663 rows=1220793 loops=1)
         Index Cond: (title ~~* 'x264'::text)
 Planning time: 1.168 ms
 Execution time: 1755.944 ms

On Sun, Nov 13, 2016 at 10:04 PM, Oleg Bartunov <obartunov@gmail.com> wrote:
>
>
> On Sun, Nov 13, 2016 at 2:54 PM, Aaron Lewis <the.warl0ck.1989@gmail.com>
> wrote:
>>
>> I have a simple table with Trigram index,
>>
>> create table mytable(hash char(40), title text);
>> create index title_trgm_idx on mytable using gin(title gin_trgm_ops);
>>
>> When I run a query with 10m rows, it uses the Trigram index, but takes
>> 3s to execute, very slow.
>> (I have 80m rows, but only inserted 10m for testing purpose)
>>
>> test=# select count(*) from mytable;
>>   count
>> ----------
>>  13971887
>> (1 row)
>>
>> test=# explain analyze select * from mytable where title ilike 'x264';
>>                                                               QUERY PLAN
>>
>> --------------------------------------------------------------------------------------------------------------------------------------
>>  Bitmap Heap Scan on mytable  (cost=462.69..5639.67 rows=1380
>> width=83) (actual time=2937.308..2937.308 rows=0 loops=1)
>>    Recheck Cond: (title ~~* 'x264'::text)
>>    Rows Removed by Index Recheck: 11402855
>>    Heap Blocks: exact=39557 lossy=158010
>>    ->  Bitmap Index Scan on title_trgm_idx  (cost=0.00..462.35
>> rows=1380 width=0) (actual time=342.440..342.440 rows=1220793 loops=1)
>>          Index Cond: (title ~~* 'x264'::text)
>>  Planning time: 0.611 ms
>>  Execution time: 2937.729 ms
>> (8 rows)
>>
>> Any ideas to speed things up?
>
>
>    Rows Removed by Index Recheck: 11402855
>    Heap Blocks: exact=39557 lossy=158010
>
> You need to increase work_mem
>>
>>
>> --
>> Best Regards,
>> Aaron Lewis - PGP: 0x13714D33 - http://pgp.mit.edu/
>> Finger Print:   9F67 391B B770 8FF6 99DC  D92D 87F6 2602 1371 4D33
>>
>>
>> --
>> Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
>> To make changes to your subscription:
>> http://www.postgresql.org/mailpref/pgsql-general
>
>



--
Best Regards,
Aaron Lewis - PGP: 0x13714D33 - http://pgp.mit.edu/
Finger Print:   9F67 391B B770 8FF6 99DC  D92D 87F6 2602 1371 4D33

I have a simple table with Trigram index,

create table mytable(hash char(40), title text);
create index title_trgm_idx on mytable using gin(title gin_trgm_ops);

When I run a query with 10m rows, it uses the Trigram index, but takes
3s to execute, very slow.
(I have 80m rows, but only inserted 10m for testing purpose)

test=# select count(*) from mytable;
  count
----------
 13971887
(1 row)

test=# explain analyze select * from mytable where title ilike 'x264';
                                                              QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------
 Bitmap Heap Scan on mytable  (cost=462.69..5639.67 rows=1380
width=83) (actual time=2937.308..2937.308 rows=0 loops=1)
   Recheck Cond: (title ~~* 'x264'::text)
   Rows Removed by Index Recheck: 11402855
   Heap Blocks: exact=39557 lossy=158010
   ->  Bitmap Index Scan on title_trgm_idx  (cost=0.00..462.35
rows=1380 width=0) (actual time=342.440..342.440 rows=1220793 loops=1)
         Index Cond: (title ~~* 'x264'::text)
 Planning time: 0.611 ms
 Execution time: 2937.729 ms
(8 rows)

Any ideas to speed things up?