[one-users] Need some details on CPU/vCPU in Sunstone and capacity planning.

[Carlos Martín Sánchez]

Hi,

On Fri, Feb 21, 2014 at 3:30 PM, kiran ranjane <kiran.ranjane at gmail.com>wrote:

> Hi Carlos,
>
> Thanks for the explanation, Few more question on this :
>
> How many vCPU a core can manage?
> What is the maximum vCPU that can be allocated to a VM?
>
> Thanks
> Kiran Ranjane
>

There isn't a limit enforced by OpenNebula. The limit depends on the
underlying hypervisor.

Regards
--
Carlos Martín, MSc
Project Engineer
OpenNebula - Flexible Enterprise Cloud Made Simple
www.OpenNebula.org <http://www.opennebula.org/> | cmartin at opennebula.org |
@OpenNebula <http://twitter.com/opennebula> <cmartin at opennebula.org>





>
> On Thu, Feb 20, 2014 at 9:32 PM, Carlos Martín Sánchez <
> cmartin at opennebula.org> wrote:
>
>> Hi,
>>
>> On Mon, Feb 17, 2014 at 4:29 PM, kiran ranjane <kiran.ranjane at gmail.com>
>>  wrote:
>>
>> Hi Everyone,
>>>
>>> I have few questions with regards to cpu and vcpu while creating
>>> templates in Sunstone :
>>>
>>> 1) What is the difference between cpu and vcpu according to opennebula
>>> architecture?
>>>
>>
>> CPU refers to the physical Host cpu cores, the one you see in the
>> "allocated cpu" column.
>> vCPU is simply the number of cpus that the VM guest will see, but it does
>> not affect the physical resources assigned to the VM.
>>
>> 2) How do we calculate vCPU, For example if I have 1 processor with 8
>>> core then how much vCPU can I allocate.
>>>
>>
>> They are not related. But the scheduler will not deploy more VMs when the
>> sum of the running VM CPU reaches 8.
>>
>>
>>> 3) How can I manage cpu capacity (recommendation) in a better way so
>>> that I can utilize it in optimum way.
>>>
>> 4) Any calculation on CPU or formula that we need to use before assigning
>>> it to Virtual machine, For example if I use 1 CPU in the virtual machine it
>>> uses 1 core of the processor? is this correct.
>>> 5) What is the use of vCPU, If I have allocated 1 CPU and 2 vCPU to the
>>> virtual machine thus that mean it will use 1 core of the physical and
>>> inside the VM it will show 2 cpu assigned to it?
>>>
>>
>> The purpose of vCPU is to allow cpu overcommitment. For example, for a
>> host with 8 cores, you could create 4 VMs with
>> cpu=2, vcpu=2
>>
>> But if your VMs workload is not very cpu-intensive, you could decide to
>> launch 16 VMs
>> cpu=0.5, vcpu=2
>>
>>
>> To ensure that the VMs are not using more CPU than the amount assigned,
>> you may need to configure your hypervisor. For example, in kvm you need to
>> configure cgroups [1]
>>
>> Regards
>>
>> [1]
>> http://docs.opennebula.org/stable/administration/virtualization/kvmg.html#working-with-cgroups-optional
>> --
>> Carlos Martín, MSc
>> Project Engineer
>> OpenNebula - Flexible Enterprise Cloud Made Simple
>> www.OpenNebula.org | cmartin at opennebula.org | @OpenNebula<http://twitter.com/opennebula>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20140224/27e2b988/attachment-0002.htm>