Hi,
On 12 August 2011 14:57, Greg Smith g...@2ndquadrant.com wrote:
I'm about to evaluate this SSD card: FusionIO ioDrive Duo [1]. The
main reason for this experiment is to see if SSD can significantly
improve query performance
The result is that FusionIO will help to our queries which was
Hi,
On 12 August 2011 14:57, Greg Smith g...@2ndquadrant.com wrote:
ioDrive hardware is fast at executing all sorts of I/O, but it particularly
excels compared to normal drives with really random workloads.
That's what I hope for :). It looks like that ioDrive is 3 to 5 times
faster for seq IO
2011/8/10 Ondrej Ivanič ondrej.iva...@gmail.com:
Ups! Well spotted Tomas! The actual values are:
random_page_cost = 2
seq_page_cost = 1
With the SSD I would set these to the same value of 1. That's what I do.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make
On 08/12/2011 04:24 PM, Vick Khera wrote:
2011/8/10 Ondrej Ivaničondrej.iva...@gmail.com:
Ups! Well spotted Tomas! The actual values are:
random_page_cost = 2
seq_page_cost = 1
With the SSD I would set these to the same value of 1. That's what I do.
That probably makes sense
There have been several discussions for SSD in recent months although not
specific to Fusion IO drives.
See http://archives.postgresql.org/pgsql-general/2011-04/msg00460.php . You
can search the archives for more such reference.
Amitabh
2011/8/11 Ondrej Ivanič ondrej.iva...@gmail.com
Hi,
Hi,
2011/8/11 Amitabh Kant amitabhk...@gmail.com:
There have been several discussions for SSD in recent months although not
specific to Fusion IO drives.
See http://archives.postgresql.org/pgsql-general/2011-04/msg00460.php . You
can search the archives for more such reference.
I've read
On 08/09/2011 07:17 PM, Ondrej Ivanič wrote:
I'm about to evaluate this SSD card: FusionIO ioDrive Duo [1]. The
main reason for this experiment is to see if SSD can significantly
improve query performance
Database size is around ~1.4TB. Main tables occupied around 1/3
(450GB, ~220mil rows) and
On 10 Srpen 2011, 1:17, Ondrej Ivanič wrote:
- What needs to be changed at Postgres/Operating system level? The
obvious one is to change random_page_cost (now: 2) and seq_page_cost
(now: 4). What else should I look at?
Are you sure about this? I'm not quite sure setting seq_page_cost=4 and
Hi,
2011/8/10 Tomas Vondra t...@fuzzy.cz:
On 10 Srpen 2011, 1:17, Ondrej Ivanič wrote:
- What needs to be changed at Postgres/Operating system level? The
obvious one is to change random_page_cost (now: 2) and seq_page_cost
(now: 4). What else should I look at?
Are you sure about this? I'm
Hi,
I'm about to evaluate this SSD card: FusionIO ioDrive Duo [1]. The
main reason for this experiment is to see if SSD can significantly
improve query performance. So, I have the following questions:
- Could you please share your experience with SSD? Any issues?
- What needs to be changed at
10 matches
Mail list logo