openstack team mailing list archive
-
openstack team
-
Mailing list archive
-
Message #12547
Re: Greatest deployment?
On Thu, May 31, 2012 at 3:54 AM, Matt Joyce <matt.joyce@xxxxxxxxxxxxxxxx> wrote:
> On Wed, May 30, 2012 at 12:39 PM, Soren Hansen <soren@xxxxxxxxxxx> wrote:
>>
>> 2012/5/30 Matt Joyce <matt.joyce@xxxxxxxxxxxxxxxx>:
>> > Secondly, while LXC does provide a lot of native access, it still does
>> > paging management internally just as kvm does. So direct memory
>> > management
>> > ( some HPC users like this ) becomes just as problematic as it is in
>> > kvm.
>> > Lots of overhead.
>>
>> I'm not convinced this is accurate. Can you provide some kind of
>> reference for this?
>>
>
> Okay so KVM uses a nastier abstraction layer in the form of shadow paging,
> while LXC simply relies on cgroups for memory isolation. Obviously two very
> different beasts. But there is the overhead of cgroup accounting and
> resource management inside LXC. It's not the same, and not nearly as much
> overhead but it's still there.
Doesn't hardware feature such as Extend-Page-Table (EPT) or
Nested-Page-Table (NPT) help in the case of KVM? There's still some
extra cost than native page table but should be better than software
solution in most use cases.
> pick your mirror: /kernel/Documentation/cgroups/memory.txt
>
> would be the best docs I know of.
>
> -Matt
>
> _______________________________________________
> Mailing list: https://launchpad.net/~openstack
> Post to : openstack@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~openstack
> More help : https://help.launchpad.net/ListHelp
>
--
Regards
Huang Zhiteng
References