inside the vms guests.
I'll try to post my results next month
- Mail original -
De: "Andrew Thrift"
À: "aderumier"
Cc: "dietmar" , "pve-devel"
Envoyé: Vendredi 30 Janvier 2015 04:52:45
Objet: Re: [pve-devel] [PATCH] iothreads : create 1 i
; À: "aderumier"
> Cc: "pve-devel"
> Envoyé: Jeudi 29 Janvier 2015 08:54:43
> Objet: Re: [pve-devel] [PATCH] iothreads : create 1 iothread by virtio
> device
>
> > I going to have a full ssd ceph cluster next month (18x s 3500 on 3
> hosts),
> > wi
l"
Envoyé: Jeudi 29 Janvier 2015 08:54:43
Objet: Re: [pve-devel] [PATCH] iothreads : create 1 iothread by virtio device
> I going to have a full ssd ceph cluster next month (18x s 3500 on 3 hosts),
> with more powerfull nodes and more powerfull clients (20 cores 3,1Gghz)
>
>
> I going to have a full ssd ceph cluster next month (18x s 3500 on 3 hosts),
> with more powerfull nodes and more powerfull clients (20 cores 3,1Gghz)
>
> I'll do benchmarks and post results with differents setup.
Would be interesting to additionally benchmark with sheepdog, so that we can
compa
help.
Alexandre
(I'm also going to try to test vhost-scsi feature of qemu soon, which is some
kind of full passthrough, bypassing all emulation layers)
- Mail original -
De: "Andrew Thrift"
À: "aderumier"
Envoyé: Jeudi 29 Janvier 2015 02:55:11
Objet: Re: [pv
> The patch series for block jobs was here
> http://lists.gnu.org/archive/html/qemu-devel/2014-10/msg02191.html
>
>
> (I'm not sure about the difference between proxmox backup patches and qemu
> implemention)
We have a bunch of patches above that (vma).
_
.org/archive/html/qemu-devel/2014-10/msg02191.html
(I'm not sure about the difference between proxmox backup patches and qemu
implemention)
- Mail original -
De: "dietmar"
À: "aderumier"
Cc: "pve-devel"
Envoyé: Jeudi 15 Janvier 2015 17:13:35
Obje
> I'll try to send a patch series tomorrow,
>
> I'll also implement iothreads for virtio-scsi && multiqueus for virtio-scsi.
Besides, we also need to rework all our qemu patches, to make
them work with iothreads.
___
pve-devel mailing list
pve-devel@
Ok,
I'll try to send a patch series tomorrow,
I'll also implement iothreads for virtio-scsi && multiqueus for virtio-scsi.
- Mail original -
De: "dietmar"
À: "aderumier" , "pve-devel"
Envoyé: Jeudi 15 Janvier 2015 09:51:54
Objet: Re: [
> I'm also thinking to improve this patch,
>
> with settings iothread as a drive property instead to enable it globally.
> Like this we can enable it only on drive which need best performance
>
> (I think that nobody use currently the iothread option)
>
> What do you think about this ?
Yes, lo
I'm also thinking to improve this patch,
with settings iothread as a drive property instead to enable it globally.
Like this we can enable it only on drive which need best performance
(I think that nobody use currently the iothread option)
What do you think about this ?
- Mail original
for better performance, we create 1 iothread "iothread-deviceid" by virtio
device.
we can hot-add|del iothread with object_add|del qmp command
Benched with rbd : I can reach around 4iops by iothreads
Signed-off-by: Alexandre Derumier
---
PVE/QemuServer.pm | 47 ++
12 matches
Mail list logo
