> >>I though it apply globally, but I am unsure now.
>
> Just tested, you are right, it's globally for the number of cpus in the groups
> (default all cpus).
>
> so , with 20cores host, and cpus limit 2,
>
> you can indeed use 2 cpu full, or split the load accross all cpus.
>
> (It's just a c
- Mail original -
De: "dietmar"
À: "aderumier"
Cc: "pve-devel"
Envoyé: Samedi 30 Mai 2015 19:26:54
Objet: Re: [pve-devel] qemu-server : cgroups implementation through systemd-run
> But maybe I'm wrong,
>
> I don't know if cfs_quota_us valu
> But maybe I'm wrong,
>
> I don't know if cfs_quota_us value apply globally on all cpus , or
> individually on each cpu.
I though it apply globally, but I am unsure now.
Anyway, I guess I will rename 'cpus' into 'cpulimit', and we can then add
further
cpu restrictions in future.
__
ame config with settings 2cpu set.
1 monothreaded application, will use 100% of 1core.
So 10x faster than previously
That's an huge difference.
- Mail original -
De: "dietmar"
À: "aderumier"
Cc: "pve-devel"
Envoyé: Samedi 30 Mai 2015 12:49:1
ation, will use 100% of 1core.
So 10x faster than previously
That's an huge difference.
- Mail original -
De: "dietmar"
À: "aderumier"
Cc: "pve-devel"
Envoyé: Samedi 30 Mai 2015 12:49:19
Objet: Re: [pve-devel] qemu-server : cgroups implemen
> >>You think I should call it 'cpulimit' instead of 'cpus'?
>
> I mean, if you go in a container, do a "top", you'll always see the full
> number of cpus of the host.
>
> in openvz :
> cfs_quota_us = cpulimit.
> cpus = number of cores assign in a container.
>
>
> It's possible to really fix
er of cpus inside the container with
cpu.cpuset.
(The only problem is that we need to specify cpuids we want to assign, so it's
not easy to have a good auto loadbalacing with multiple containers)
- Mail original -
De: "dietmar"
À: "aderumier"
Cc: "pve-d
> >>Great. I guess we can also hotplug that?
>
> Yes sure. it's just a write in /sys/fs/cgroups.
>
> This could be done online too in lxc.
OK
> BTW, I see that in lxc gui, the cpus number field =
> lxc.cgroup.cpu.cfs_quota_us, so it's more cpulimit.
> (all cores are present in container)
Sor
-- Mail original -
De: "dietmar"
À: "aderumier"
Cc: "pve-devel"
Envoyé: Vendredi 29 Mai 2015 11:36:50
Objet: Re: [pve-devel] qemu-server : cgroups implementation through systemd-run
> >>Answering myself, I found we can set this with
> >>
> >>Answering myself, I found we can set this with
> >>
> >>CPUQuota=XX%
> >>
> >>So we can directly map this to the 'cpulimit' parameter?
>
> yes indeed. I'll send a patch monday for this.
Great. I guess we can also hotplug that?
___
pve-devel mail
" , "pve-devel"
Envoyé: Vendredi 29 Mai 2015 08:38:09
Objet: Re: [pve-devel] qemu-server : cgroups implementation through systemd-run
Answering myself, I found we can set this with
CPUQuota=XX%
So we can directly map this to the 'cpulimit' parameter?
On 05/
Answering myself, I found we can set this with
CPUQuota=XX%
So we can directly map this to the 'cpulimit' parameter?
On 05/29/2015 08:33 AM, Dietmar Maurer wrote:
Are there other things we can set this way?
For example can we limit CPU with cfs_period_us and cfs_quota_us?
I guess that woul
Are there other things we can set this way?
For example can we limit CPU with cfs_period_us and cfs_quota_us?
I guess that would be interesting for hosters.
On 05/28/2015 03:59 PM, Alexandre Derumier wrote:
I have tested it, it's working fine.
(I don't known how HA will be managed but it sho
applied, thanks!
On 05/28/2015 03:59 PM, Alexandre Derumier wrote:
I have tested it, it's working fine.
(I don't known how HA will be managed but it should work too)
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/
I have tested it, it's working fine.
(I don't known how HA will be managed but it should work too)
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
15 matches
Mail list logo
