Re: BUG #6186: out of memory while analyze - Mailing list pgsql-bugs

From Lampa
Subject Re: BUG #6186: out of memory while analyze
Date
Msg-id CAG_XuNWnp=ZC4ErD+zW1bF_vPcX5WhWCa78f4D50xg8asBXhRQ@mail.gmail.com
Whole thread Raw
In response to Re: BUG #6186: out of memory while analyze  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: BUG #6186: out of memory while analyze  (Guillaume Smet <guillaume.smet@gmail.com>)
List pgsql-bugs
Hello,

ok or write that there is limit for 32bit system.

With default value 100 have bad perfomance :-(

But thank you for yours time.

2011/8/31 Tom Lane <tgl@sss.pgh.pa.us>:
> Lampa <lampacz@gmail.com> writes:
>> default_statistics_target =3D 8000 =C2=A0 =C2=A0 =C2=A0 =C2=A0# range 1-=
10000
>
>> Problem seems to be in default_statistics_target values from approx
>> 6300. Up to 6300 analyze works but on another machine (64bit) works
>> 10000 without problems on same but larger table (more data)
>
> Well, there's a reason why the default value of that parameter is 100
> and not 10000 ;-). =C2=A0The larger it is, the more memory space it's goi=
ng
> to take to do an ANALYZE. =C2=A0I don't see any bug here, you're just try=
ing
> to do more than you can do in a 32-bit machine.
>
> I guess we could improve the documentation of default_statistics_target
> to point out that it costs not only more time but more memory in
> ANALYZE.
>
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0regards, tom lane
>



--=20
Lampa

pgsql-bugs by date:

Previous
From: "Mark Lamberton"
Date:
Subject: BUG #6191: One click installer fails
Next
From: Guillaume Smet
Date:
Subject: Re: BUG #6186: out of memory while analyze