Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Hanno Hirschberger
Hi Martin, On 23.09.2015 10:51, Martin Truhlář wrote: Tests revealed, that problem is somewhere in disk array itself. are you familiar with the ashift problem on 4k drives? My best guess would be that the 1 TB WD drives are emulating a block size of 512 bytes while using 4k sectors

Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Steffen Wagner
Hi Hanno, how do you calculate the best ashift value? Thanks, Steffen -Original Message- From: OmniOS-discuss [mailto:omnios-discuss-boun...@lists.omniti.com] On Behalf Of Hanno Hirschberger Sent: Mittwoch, 23. September 2015 14:43 To: omnios-discuss@lists.omniti.com Subject: Re:

Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Michael Rasmussen
Have you tried running iperf? On September 23, 2015 4:06:53 PM CEST, "Martin Truhlář" wrote: >Hi Hanno > >Thank you for your advice, unfortunatelly on dpool is ashift already >set to 12 without any impact on performance. > >Martin > >dpool: >version: 5000 >

Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Guenther Alka
Poor write performance is often related to sync write. Enable write back cache for your logical units (and disable ZFS sync property on the filesystem for filebased lu's) and redo some performance tests. Gea Am 23.09.2015 um 10:51 schrieb Martin Truhlář: Tests revealed, that problem is

Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Stephan Budach
Am 23.09.15 um 10:51 schrieb Martin Truhlář: Tests revealed, that problem is somewhere in disk array itself. Write performance of disk connected directly (via iSCSI) to KVM is poor as well, even write performance measured on Omnios is very poor. So loop is tightened, but there still remains

Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Martin Truhlář
Hi Hanno Thank you for your advice, unfortunatelly on dpool is ashift already set to 12 without any impact on performance. Martin dpool: version: 5000 name: 'dpool' state: 0 txg: 423442 pool_guid: 8301756920046328435 hostid: 390978448 hostname: 'archnas'

Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Michael Rasmussen
On Wed, 23 Sep 2015 17:23:24 +0200 Stephan Budach wrote: > > At any way, you will need to get the performance of your zpools straight > first, before even beginning to think on how to tweak the performance over > the network. > Since his pool is comprised of vdev

Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Stephan Budach
Am 23.09.15 um 18:59 schrieb Michael Rasmussen: On Wed, 23 Sep 2015 17:23:24 +0200 Stephan Budach wrote: At any way, you will need to get the performance of your zpools straight first, before even beginning to think on how to tweak the performance over the network.

Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Martin Truhlář
Tests revealed, that problem is somewhere in disk array itself. Write performance of disk connected directly (via iSCSI) to KVM is poor as well, even write performance measured on Omnios is very poor. So loop is tightened, but there still remains lot of possible hacks. I strived to use

Re: [OmniOS-discuss] iSCSI poor write performance

2015-09-23 Thread Michael Rasmussen
On Wed, 23 Sep 2015 19:56:26 +0200 Stephan Budach wrote: > Huh? Where did that escape me? I don't think, that the pool layout showed any > remote disks, they all Sorry, reading to hastig. Misread phys_path: '/scsi_vhci/disk@g50014ee00400fa16:a' for iscsi. --