Personaly I would split into two RAID 1s. One for pg_xlog, one for
the rest. This gives probably the best performance/reliability
combination.
Alex.
On 12/10/05, Carlos Benkendorf <[EMAIL PROTECTED]> wrote:
> Hello,
>
> I would like to know which is the best configuration to use 4 scsi drives
>
On 12/12/05, Tom Lane <[EMAIL PROTECTED]> wrote:
Mike C <[EMAIL PROTECTED]> writes:> CLUSTER on PC_TRAFFIC_IDX3 gives me significantly improved performance:How can you tell? Neither of these are EXPLAIN ANALYZE output.
regards,
tom lane
Sorry that's a result of my bad reco
Mike C <[EMAIL PROTECTED]> writes:
> CLUSTER on PC_TRAFFIC_IDX3 gives me significantly improved performance:
How can you tell? Neither of these are EXPLAIN ANALYZE output.
regards, tom lane
---(end of broadcast)---
TIP 5: d
Hello,I've got a table with ~60 Million rows and am having
performance problems querying it. Disks are setup as 4x10K SCSI 76GB,
RAID 1+0. The table is being inserted into multiple times every second
of the day, with no updates and every 2nd day we delete 1/60th of the
data (as it becomes old). Vac
Paal,
> I'm currently benchmarking several RDBMSs with respect to
> analytical query performance on medium-sized multidimensional
> data sets. The data set contains 30,000,000 fact rows evenly
> distributed in a multidimensional space of 9 hierarchical
> dimensions. Each dimension has 8000 me
=?iso-8859-1?Q?P=E5l_Stenslet?= <[EMAIL PROTECTED]> writes:
> I have established similar conditions for the query in PostgreSQL, and =
> it runs in about 30 seconds. Again the CPU utilization is high with no =
> noticable I/O. The query plan is of course very different from that of =
> Oracle, sinc
Edison Azzi <[EMAIL PROTECTED]> writes:
> You are rigth, the planner will not eliminate the join, see:
> select * from cta_pag a, cta_pag p where a.nrlancto=p.nrlancto and
> p.nrlancto = 21861;
> EXPLAIN:
> Nested Loop (cost=0.00..11.48 rows=1 width=816)
> -> Index Scan using cta_pag_pk on c
On Sun, Dec 11, 2005 at 11:53:36AM +, Carlos Benkendorf wrote:
> I would like to use autovacuum but is not too much expensive
> collecting row level statistics?
The cost depends on your usage patterns. I did tests with one of
my applications and saw no significant performance difference for
s
Hi all,
Thanks for the reply. I made some more test and find out that the
problem is with the <<= operator for the network type. Can I create
index which to work with <<=. Because if I use = the index is used. But
not for <<=.
iplog=# explain analyze SELECT *
iplog-#
I'm currently benchmarking several RDBMSs with respect to
analytical query performance on medium-sized multidimensional data sets. The
data set contains 30,000,000 fact rows evenly distributed in a multidimensional
space of 9 hierarchical dimensions. Each dimension has 8000 members.
The
Richard Huxton escreveu:
Edison Azzi wrote:
Hi,
I´m trying to optimize some selects between 2 tables and the best way
I found was
alter the first table and add the fields of the 2nd table. I adjusted
the contents and
now a have only one table with all info that I need. Now resides my
probl
> Bruce Momjian writes:
> > Tom Lane wrote:
> >> 1. You don't want number of clients (-c) much higher than scaling factor
> >> (-s in the initialization step).
>
> > Should we throw a warning when someone runs the test this way?
>
> Not a bad idea (though of course only for the "standard" script
Hi, I would like to use autovacuum but is not too much expensive collecting row level statistics? Are there some numbers that I could use? Thanks in advance! Benkendorf
Yahoo! doce lar. Faça do Yahoo! sua homepage.
13 matches
Mail list logo
