| From: | Hervé Piedvache <footcow(at)noos(dot)fr> |
|---|---|
| To: | pgsql-performance(at)postgresql(dot)org |
| Cc: | Josh Berkus <josh(at)agliodbs(dot)com>, Hervé Piedvache <herve(at)elma(dot)fr> |
| Subject: | Re: TSearch2 and optimisation ... |
| Date: | 2004-08-26 19:30:40 |
| Message-ID: | 200408262130.40500.footcow@noos.fr |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
Le Jeudi 26 Août 2004 19:48, Josh Berkus a écrit :
> Herve'
>
> > (cost=0.00..4052.84 rows=1351 width=166) (actual time=109.766..5415.108
> > rows=139 loops=1)
> > Index Cond: (idxfti @@ '\'postgresql\''::tsquery)
> > Filter: (idxfti @@ '\'postgresql\''::tsquery)
> >
> >From this, it looks like your FTI index isn't fitting in your sort_mem.
>
> What's sort_mem at now? Can you increase it?
shared_buffers = 3000
sort_mem = 10240
> Overall, though, I'm not sure you can get this sub-1s without a faster
> machine. Although I'm doing FTI on about 25MB of FTI text on a
> single-processor machine, and getting 40ms response times, so maybe we can
> ...
Sorry I missed understand what you mean here ...
You tell me to upgrade the hardware but you manage a 25 Mb with a single
processor ?? What you mean ?
My database is about 450 Mb ...
Regards,
--
Bill Footcow
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Josh Berkus | 2004-08-26 19:48:25 | Re: Equivalent praxis to CLUSTERED INDEX? |
| Previous Message | Magnus Hagander | 2004-08-26 19:30:24 | Re: Equivalent praxis to CLUSTERED INDEX? |