Re: GIN index creation extremely slow ? - Mailing list pgsql-hackers

From Stefan Kaltenbrunner
Subject Re: GIN index creation extremely slow ?
Date
Msg-id 44A1834A.40107@kaltenbrunner.cc
Whole thread Raw
In response to Re: GIN index creation extremely slow ?  (Teodor Sigaev <teodor@sigaev.ru>)
Responses Re: GIN index creation extremely slow ?  (Teodor Sigaev <teodor@sigaev.ru>)
List pgsql-hackers
Teodor Sigaev wrote:
>> test=# CREATE INDEX idxFTI_idx ON wikipedia USING gist(vector);
>> CREATE INDEX
>> Time: 416122.896 ms
>>
>> so about 7 minutes - sounds very reasonable
>>
>> test=# CREATE INDEX idxFTI2_idx ON wikipedia USING gin(vector);
>> CREATE INDEX
>> Time: 52681605.101 ms
> 
> I'll look at this,  but GiST time creation is suspiciously small.
> Can you test on smaller table, for example with 100000 records and if
> results are repeat, pls, send to me test suite...

I won't have access to the original testcase and server for a few days
but I just redid some testing on a slower personal box of mine with a
smaller(but similiar) testset and on that box I could not reproduce that
issue.
So the problem is either caused by the size of the table or somehow by
the data itself :-(


Stefan


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: posix_fadvise versus old kernels
Next
From: Tom Lane
Date:
Subject: Fully replacing ps_status (was Re: [COMMITTERS] pgsql: Add GUC update_process_title to control whether 'ps' display is)