[occi-wg] Scheduling parameters
Ignacio Martin Llorente
llorente at dacya.ucm.es
Wed Apr 15 06:09:56 CDT 2009
About OpenNebula and Cloud infrastructures:
- OpenNebula can be primarily used as a virtualization tool to manage
your virtual infrastructure in the data-center or cluster. You know,
this application is usually referred as private cloud,
- With OpenNebula you can also dynamically scale it to multiple
external clouds, so building a hybrid cloud.
- When combined with a Cloud interface (Eucalyptus, Nimbus...) ,
OpenNebula can be used as engine for public clouds, providing a
scalable and dynamic management of the back-end infrastructure.
You can also use OpenNebula as Cloud Broker. In fact we are now
executing some use cases across EC2 and ElasticHosts sites, without
using local infrastructure. In this specific case, Reservoir is doing
research on cross-site VM placement heuristics and algorithms
different categories of parameters:
–Local and remote resource availability
–Negotiation and placement decision time
Ignacio M. Llorente, Full Professor (Catedratico): web http://dsa-research.org/llorente
and blog http://imllorente.dsa-research.org/
DSA Research Group: web http://dsa-research.org and blog http://blog.dsa-research.org
Globus GridWay Metascheduler: http://www.GridWay.org
OpenNebula Virtual Infrastructure Engine: http://www.OpenNebula.org
On 15/04/2009, at 12:56, Mehdi Sheikhalishahi wrote:
> Dear All:
> I need some information about the back-end part of current
> Metascheduling for Clouds to provide some useful use-cases. In fact,
> at first, Is there any Metascheduler for Clouds?
> Can we classify OpenNebula as a Meta-scheduler for Clouds?
> If yes, what is its current policies, algorithms and optimization for
> Metascheduling in Clouds?
> On Tue, Apr 14, 2009 at 6:06 PM, Edmonds, AndrewX
> <andrewx.edmonds at intel.com> wrote:
>> Thanks Andre - I'll get around at some stage to supply some
>> feedback to the GLUE WG, but EU deliverables take precedence for
>> the moment ;-)
>> I'd also agree on your point of requiring some means of resource
>> -----Original Message-----
>> From: Andre Merzky [mailto:andremerzky at gmail.com] On Behalf Of
>> Andre Merzky
>> Sent: 14 April 2009 17:02
>> To: Sam Johnston
>> Cc: Edmonds, AndrewX; Andre Merzky; occi-wg at ogf.org
>> Subject: Re: [occi-wg] Scheduling parameters
>> Hi Sam, Andrew,
>> as Alexander and Alexis in the other branch of this thread
>> (about scheduling), I just wanted to make sure that other
>> OGF groups and standards get considered, and involved, *when
>> appropriate*. So I'm happy to hear that people have
>> considered to use GLUE, and that GLUE MAY be usable as
>> alternative representation, etc.
>> Quoting [Sam Johnston] (Apr 14 2009):
>>> Further to Andrew's comments, I've thus far tried to avoid fixed
>>> schemas for anything, preferring tags and attributes and
>>> deferring a
>>> lot of that detail to supporting standards like OVF. I don't
>>> see a
>>> problem with using GLUE as an "alternate" representation and/or
>>> to it using <link>s or an extension.
>>> I hope that answers your question... basically I suggest that we
>>> [re]use it if we need to but not before.
>>> On Tue, Apr 14, 2009 at 2:36 PM, Edmonds, AndrewX
>>> <andrewx.edmonds at intel.com> wrote:
>>>> Hey Andre - we reviewed GLUE in SLA at SOI to see if it would
>>>> meet our
>>>> needs for infrastructure provisioning but for the most our
>>>> feelings are that it wouldn't suit. Would you have any other
>>>> viewpoint on this?
>> Andrew, would you be able to feed back to the GLUE WG where
>> you considered the standarde to fall short of your
>> requirements? They might be interested to learn about that.
>>> But anyway: do you expect GLUE to play any role in respect to the
>>> specification of (VM) resource requirements?
>> I am actually not familiar enough with the topic to really
>> have an educated opinion. As an observer, it seems that the
>> OCCI will need to touch resource description at some point
>> or the other, to specify requirements to a VM for example,
>> and GLUE seems to aim at that type of use case. Not sure if
>> any from the GLUE people are listening here. If not, it
>> might be worthewhile to get their feedback at next OGF, by
>> going to their session...
>> Cheers, Andre.
>> Nothing is ever easy.
>> Intel Ireland Limited (Branch)
>> Collinstown Industrial Park, Leixlip, County Kildare, Ireland
>> Registered Number: E902934
>> This e-mail and any attachments may contain confidential material for
>> the sole use of the intended recipient(s). Any review or distribution
>> by others is strictly prohibited. If you are not the intended
>> recipient, please contact the sender and delete all copies.
>> occi-wg mailing list
>> occi-wg at ogf.org
> Best Regards,
> Mehdi Sheikhalishahi
> occi-wg mailing list
> occi-wg at ogf.org
More information about the occi-wg