Re: [PERFORM] Revisiting disk layout on ZFS systems

2014-05-01 Thread Josh Berkus
On 04/28/2014 08:47 AM, Karl Denninger wrote: > The odd thing is that I am getting better performance with a 128k record > size on this application than I get with an 8k one! Not only is the > system faster to respond subjectively and can it sustain a higher TPS > load objectively but the I/O busy

Re: [PERFORM] Revisiting disk layout on ZFS systems

2014-04-29 Thread Karl Denninger
On 4/29/2014 3:13 AM, Albe Laurenz wrote: Karl Denninger wrote: I've been doing a bit of benchmarking and real-world performance testing, and have found some curious results. [...] The odd thing is that I am getting better performance with a 128k record size on this application than I get wi

Re: [PERFORM] Revisiting disk layout on ZFS systems

2014-04-29 Thread Albe Laurenz
Karl Denninger wrote: > I've been doing a bit of benchmarking and real-world performance > testing, and have found some curious results. [...] > The odd thing is that I am getting better performance with a 128k record > size on this application than I get with an 8k one! [...] > What I am curio

Re: [PERFORM] Revisiting disk layout on ZFS systems

2014-04-28 Thread Karl Denninger
On 4/28/2014 1:26 PM, Jeff Janes wrote: On Mon, Apr 28, 2014 at 11:07 AM, Karl Denninger > wrote: Isn't WAL essentially sequential writes during normal operation? Only if you have some sort of non-volatile intermediary, or are willing to risk your data integr

Re: [PERFORM] Revisiting disk layout on ZFS systems

2014-04-28 Thread Karl Denninger
On 4/28/2014 1:22 PM, Heikki Linnakangas wrote: On 04/28/2014 09:07 PM, Karl Denninger wrote: The WAL is fsync'd frequently. My guess is that that causes a lot of extra work to repeatedly recompress the same data, or something like that. It shouldn't as ZFS re-writes on change, and what's sho

Re: [PERFORM] Revisiting disk layout on ZFS systems

2014-04-28 Thread Jeff Janes
On Mon, Apr 28, 2014 at 11:07 AM, Karl Denninger wrote: > > On 4/28/2014 1:04 PM, Heikki Linnakangas wrote: > >> On 04/28/2014 06:47 PM, Karl Denninger wrote: >> >>> What I am curious about, however, is the xlog -- that appears to suffer >>> pretty badly from 128k record size, although it compres

Re: [PERFORM] Revisiting disk layout on ZFS systems

2014-04-28 Thread Heikki Linnakangas
On 04/28/2014 09:07 PM, Karl Denninger wrote: The WAL is fsync'd frequently. My guess is that that causes a lot of extra work to repeatedly recompress the same data, or something like that. It shouldn't as ZFS re-writes on change, and what's showing up is not high I/O*count* but rather percent

Re: [PERFORM] Revisiting disk layout on ZFS systems

2014-04-28 Thread Heikki Linnakangas
On 04/28/2014 06:47 PM, Karl Denninger wrote: What I am curious about, however, is the xlog -- that appears to suffer pretty badly from 128k record size, although it compresses even more-materially; 1.94x (!) The files in the xlog directory are large (16MB each) and thus "first blush" would be t

Re: [PERFORM] Revisiting disk layout on ZFS systems

2014-04-28 Thread Karl Denninger
On 4/28/2014 1:04 PM, Heikki Linnakangas wrote: On 04/28/2014 06:47 PM, Karl Denninger wrote: What I am curious about, however, is the xlog -- that appears to suffer pretty badly from 128k record size, although it compresses even more-materially; 1.94x (!) The files in the xlog directory are l

[PERFORM] Revisiting disk layout on ZFS systems

2014-04-28 Thread Karl Denninger
I've been doing a bit of benchmarking and real-world performance testing, and have found some curious results. The load in question is a fairly-busy machine hosting a web service that uses Postgresql as its back end. "Conventional Wisdom" is that you want to run an 8k record size to match Po