On Feb 23, 2010, at 2:23 PM, Kevin Grittner wrote:
>
> Here are the values from our two largest and busiest systems (where
> we found the pg_xlog placement to matter so much). It looks to me
> like a more aggressive bgwriter would help, yes?
>
> cir=> select * from pg_stat_bgwriter ;
> -[ RECOR
Greg Smith wrote:
> You can easily quantify if the BGW is aggressive enough. Buffers
> leave the cache three ways, and they each show up as separate
> counts in pg_stat_bgwriter: buffers_checkpoint, buffers_clean
> (the BGW), and buffers_backend (the queries). Cranking it up
> further tends t
Kevin Grittner wrote:
I wonder if it might also pay to make the background writer even more
aggressive than we have, so that SELECT-only queries don't spend so
much time writing pages.
You can easily quantify if the BGW is aggressive enough. Buffers leave
the cache three ways, and they each sho
Kevin Grittner wrote:
> Alvaro Herrera wrote:
> > Actually, a transaction that performed no writes doesn't get a
> > commit WAL record written, so it shouldn't make any difference at
> > all.
>
> Well, concurrent to the web application is the replication. Would
> asynchronous commit of that po
Alvaro Herrera wrote:
> Alvaro Herrera wrote:
>> Kevin Grittner wrote:
>
>> > Anyway, given that these are replication targets, and aren't
>> > the "database of origin" for any data of their own, I guess
>> > there's no reason not to try asynchronous commit.
>>
>> Yeah; since the transactions o
Alvaro Herrera wrote:
> Kevin Grittner wrote:
> > Anyway, given that these are replication
> > targets, and aren't the "database of origin" for any data of their
> > own, I guess there's no reason not to try asynchronous commit.
>
> Yeah; since the transactions only ever write commit records to
Kevin Grittner wrote:
> Hannu Krosing wrote:
>
> > Can it be, that each request does at least 1 write op (update
> > session or log something) ?
>
> Well, the web application connects through a login which only has
> SELECT rights; but, in discussing a previous issue we've pretty well
> establ
Hannu Krosing wrote:
> Can it be, that each request does at least 1 write op (update
> session or log something) ?
Well, the web application connects through a login which only has
SELECT rights; but, in discussing a previous issue we've pretty well
established that it's not unusual for a read
Aidan Van Dyk wrote:
> Alvaro Herrera wrote:
>> Hmm, so maybe the performance benefit is not from it being on a
>> separate array, but from it being RAID1 instead of RAID5?
>
> Or the cumulative effects of:
> 1) Dedicated spindles/Raid1
> 2) More BBU cache available (I can't imagine the OS pair
* Alvaro Herrera [100211 12:58]:
> Hmm, so maybe the performance benefit is not from it being on a separate
> array, but from it being RAID1 instead of RAID5?
Or the cumulative effects of:
1) Dedicated spindles/Raid1
2) More BBU cache available (I can't imagine the OS pair writing much)
3) not be
Kevin Grittner wrote:
> Another example of why I shouldn't trust my memory. Per the
> hardware tech:
>
>
> OS: /dev/sda is RAID1 - 2 x 2.5" 15k SAS disk
> pg_xlog: /dev/sdb is RAID1 - 2 x 2.5" 15k SAS disk
>
> These reside on a ServeRAID-MR10k controller with 256MB BB cache
"Kevin Grittner" wrote:
> Jesper Krogh wrote:
>
>> Sorry if it is obvious.. but what filesystem/OS are you using and
>> do you have BBU-writeback on the main data catalog also?
> File system is xfs noatime,nobarrier for all data; OS is on ext3.
> I *think* the pg_xlog mirrored pair is hangin
On Thu, Feb 11, 2010 at 4:29 AM, Bruce Momjian wrote:
> Kevin Grittner wrote:
>> Jesper Krogh wrote:
>>
>> > Sorry if it is obvious.. but what filesystem/OS are you using and
>> > do you have BBU-writeback on the main data catalog also?
>>
>> Sorry for not providing more context.
>>
>> ATHENA:/va
Kevin Grittner wrote:
> Jesper Krogh wrote:
>
> > Sorry if it is obvious.. but what filesystem/OS are you using and
> > do you have BBU-writeback on the main data catalog also?
>
> Sorry for not providing more context.
>
> ATHENA:/var/pgsql/data # uname -a
> Linux ATHENA 2.6.16.60-0.39.3-smp
Amitabh Kant wrote:
> Just curious if you have a 16 physical CPU's or 16 cores on 4
> CPU/8 cores over 2 CPU with HT.
Four quad core CPUs:
vendor_id : GenuineIntel
cpu family : 6
model : 15
model name : Intel(R) Xeon(R) CPU X7350 @ 2.93GHz
stepping
On Tue, Feb 9, 2010 at 10:03 PM, Kevin Grittner wrote:
> Jesper Krogh wrote:
> File system is xfs noatime,nobarrier for all data; OS is on ext3. I
> *think* the pg_xlog mirrored pair is hanging off the same
> BBU-writeback controller as the big RAID, but I'd have to track down
> the hardware te
Jesper Krogh wrote:
> Sorry if it is obvious.. but what filesystem/OS are you using and
> do you have BBU-writeback on the main data catalog also?
Sorry for not providing more context.
ATHENA:/var/pgsql/data # uname -a
Linux ATHENA 2.6.16.60-0.39.3-smp #1 SMP Mon May 11 11:46:34 UTC
2009 x86
>
> Frankly, I was quite surprised by this, since some of the benchmarks
> people have published on the effects of using a separate RAID for
> the WAL files have only shown a one or two percent difference when
> using a hardware RAID controller with BBU cache configured for
> write-back.
Hi Kevin.
Due to an error during an update to the system kernel on a database
server backing a web application, we ran for a month (mid-December
to mid-January) with the WAL files in the pg_xlog subdirectory on
the same 40-spindle array as the data -- only the OS was on a
separate mirrored pair of drives. W
19 matches
Mail list logo