Re: [linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-14 Thread Dale Stephenson
> On Sep 14, 2017, at 5:00 AM, Zdenek Kabelac wrote: > > Dne 14.9.2017 v 00:39 Dale Stephenson napsal(a): >> I could create the md to use 512k chunks for RAID 0, but I wouldn’t expect >> that to have any impact on a single threaded test using 4k request size. Is >> there

Re: [linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-14 Thread Dale Stephenson
> On Sep 14, 2017, at 7:13 AM, Zdenek Kabelac wrote: > > Dne 14.9.2017 v 12:57 Gionatan Danti napsal(a): >> On 14/09/2017 11:37, Zdenek Kabelac wrote: >>> Sorry my typo here - is NOT ;) >>> >>> >>> Zdenek >> Hi Zdenek, >> as the only variable is the LVM volume type

Re: [linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-14 Thread Gionatan Danti
On 14/09/2017 11:37, Zdenek Kabelac wrote: Sorry my typo here - is NOT ;) Zdenek Hi Zdenek, as the only variable is the LVM volume type (fat/thick vs thin), why the thin volume is slower than the thick one? I mean: all other things being equal, what is holding back the thin volume?

Re: [linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-14 Thread Zdenek Kabelac
Dne 14.9.2017 v 12:57 Gionatan Danti napsal(a): On 14/09/2017 11:37, Zdenek Kabelac wrote: Sorry my typo here - is NOT ;) Zdenek Hi Zdenek, as the only variable is the LVM volume type (fat/thick vs thin), why the thin volume is slower than the thick one? I mean: all other things being

Re: [linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-14 Thread Gionatan Danti
On 14/09/2017 11:37, Zdenek Kabelac wrote: Sorry my typo here - is NOT ;) Zdenek Hi Zdenek, as the only variable is the LVM volume type (fat/thick vs thin), why the thin volume is slower than the thick one? I mean: all other things being equal, what is holding back the thin volume?

Re: [linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-14 Thread Zdenek Kabelac
Dne 14.9.2017 v 11:00 Zdenek Kabelac napsal(a): Dne 14.9.2017 v 00:39 Dale Stephenson napsal(a): On Sep 13, 2017, at 4:19 PM, Zdenek Kabelac wrote: md127 is an 8-drive RAID 0 As you can see, there’s no lvm striping; I rely on the software RAID underneath for that. 

Re: [linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-14 Thread Zdenek Kabelac
Dne 14.9.2017 v 00:39 Dale Stephenson napsal(a): On Sep 13, 2017, at 4:19 PM, Zdenek Kabelac wrote: Dne 13.9.2017 v 17:33 Dale Stephenson napsal(a): Distribution: centos-release-7-3.1611.el7.centos.x86_64 Kernel: Linux 3.10.0-514.26.2.el7.x86_64 LVM: 2.02.166(2)-RHEL7

Re: [linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-13 Thread Dale Stephenson
> On Sep 13, 2017, at 4:19 PM, Zdenek Kabelac wrote: > > Dne 13.9.2017 v 17:33 Dale Stephenson napsal(a): >> Distribution: centos-release-7-3.1611.el7.centos.x86_64 >> Kernel: Linux 3.10.0-514.26.2.el7.x86_64 >> LVM: 2.02.166(2)-RHEL7 (2016-11-16) >> Volume group consisted

Re: [linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-13 Thread Zdenek Kabelac
Dne 13.9.2017 v 17:33 Dale Stephenson napsal(a): Distribution: centos-release-7-3.1611.el7.centos.x86_64 Kernel: Linux 3.10.0-514.26.2.el7.x86_64 LVM: 2.02.166(2)-RHEL7 (2016-11-16) Volume group consisted of an 8-drive SSD (500G drives) array, plus an additional SSD of the same size. The

[linux-lvm] Performance penalty for 4k requests on thin provisioned volume

2017-09-13 Thread Dale Stephenson
Distribution: centos-release-7-3.1611.el7.centos.x86_64 Kernel: Linux 3.10.0-514.26.2.el7.x86_64 LVM: 2.02.166(2)-RHEL7 (2016-11-16) Volume group consisted of an 8-drive SSD (500G drives) array, plus an additional SSD of the same size. The array had 64 k stripes. Thin pool had -Zn option and