Search Postgresql Archives

Re: Problems with memory

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Richard Huxton wrote:
Pau Marc Munoz Torres wrote:
Hi

To your initial question all configurable options can be found and adjusted in the postgresql.conf which can be found in the top of your data dir.

You can find more detail on what the options are for at
http://www.postgresql.org/docs/8.3/interactive/runtime-config.html

Although I doubt it will fix your problem you will most likely want to tune them to get the best performance from your db once it goes live.

maybe i should give you some more explanations of my problem.

The reason for which i think that postgresql run out of memory is that: I
have a relation with 6 fields, 29 indexes and 32000 registers, the registers

So how much ram does your machine have? have you looked at how much postgres is using while it is running the queries?

What CPU's and disks/raid controller?

when i do a query as:

select * from precalc where idr(p1, p4, p6, p7, p9, 'HLA-DRB1*0101')<-2; it
works and return 128030 registers

if i do

select * from precalc where idr(p1, p4, p6, p7, p9, 'HLA-DRB1*0101')>-2;
3071970 registers, it don't work
ERROR:  relation "pssms" does not exist
CONTEXT: SQL statement "select score from PSSMS where AA= $1 and POS=1 and
MOLEC= $2 "
PL/pgSQL function "idr" line 11 at SQL statement

Start with the idr function - this is where the error comes from.
The error doesn't mention out of memory.


if i ask for explanation for both queries works:

explain will show what the planner expects to do. explain analyse will actually run through the query and give exact row counts returned etc. and will hit the error that you get when running the query.


If for that reason that i think that my machine runs out of memory, by the
way, this is not the biggest table that i have others have more than
503000000 registers, so if I try to do a cross select between tables it
could be worse.

You may hear from people on this list that have db's storing many TeraBytes of data.



--

Shane Ambler
pgSQL (at) Sheeky (dot) Biz

Get Sheeky @ http://Sheeky.Biz


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Postgresql Jobs]     [Postgresql Admin]     [Postgresql Performance]     [Linux Clusters]     [PHP Home]     [PHP on Windows]     [Kernel Newbies]     [PHP Classes]     [PHP Books]     [PHP Databases]     [Postgresql & PHP]     [Yosemite]
  Powered by Linux