From: | Craig James <craig_james(at)emolecules(dot)com> |
---|---|
To: | Josh Berkus <josh(at)agliodbs(dot)com> |
Cc: | pgsql-performance(at)postgresql(dot)org |
Subject: | Re: PostgreSQL as a local in-memory cache |
Date: | 2010-06-16 19:36:31 |
Message-ID: | 4C1927BF.9030708@emolecules.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
On 6/16/10 12:00 PM, Josh Berkus wrote:
>
>> * fsync=off => 5,100
>> * fsync=off and synchronous_commit=off => 5,500
>
> Now, this *is* interesting ... why should synch_commit make a difference
> if fsync is off?
>
> Anyone have any ideas?
I found that pgbench has "noise" of about 20% (I posted about this a couple days ago using data from 1000 identical pgbench runs). Unless you make a bunch of runs and average them, a difference of 5,100 to 5,500 appears to be meaningless.
Craig
>
>> tmpfs, WAL on same tmpfs:
>> * Default config: 5,200
>> * full_page_writes=off => 5,200
>> * fsync=off => 5,250
>> * synchronous_commit=off => 5,200
>> * fsync=off and synchronous_commit=off => 5,450
>> * fsync=off and full_page_writes=off => 5,250
>> * fsync=off, synchronous_commit=off and full_page_writes=off => 5,500
>
> So, in this test, it seems like having WAL on tmpfs doesn't make a
> significant difference for everything == off.
>
> I'll try running some tests on Amazon when I have a chance. It would be
> worthwhile to get figures without Python's "ceiling".
>
From | Date | Subject | |
---|---|---|---|
Next Message | Pierre C | 2010-06-16 20:40:49 | Re: PostgreSQL as a local in-memory cache |
Previous Message | Scott Marlowe | 2010-06-16 19:34:43 | Re: Confirm calculus |