← Back to team overview

openstack team mailing list archive

Re: distributed and heterogeneous schedulers

 

> From: Soren Hansen [soren@xxxxxxxxxxxxx]
>> 2. the current load the host is under
> 
> I still question the usefulness of nr. 2. A host that is almost
> completely idle right now might be under tremendous pressure a minute
> from now and vice versa. Even if we had useful statistics (and trend
> analysis) of the existing VM's on the host, we have no data at all on
> the requirements of the instance we're about to launch.
> 
> It seems to me that the current load only affects the time it will
> take to spin up the new instance, and as such says nothing about its
> long term (or even medium or short term!) suitability as the new
> instance's home.

True. The "last-known-load" is just a suggestion for the routing, the scheduler can choose to use it or not. 

As with all scheduling, it will only *attempt* to create the instance on that host. But if it fails it will have list of candidates to try in weighted order. Moreso, this weighting will span zones ... not just the hosts in a single zone. That's the stuff we're working on now in the distributed-scheduler branch. 

We do, however, know the requirements of the instance we're about to launch based on the Flavor. This gives us bandwidth, memory and cpu limits which we can enforce in the host. Let the hypervisor handle the rest. 

We can also add "fudge factors" to the capabilities to leave ceiling space for misbehaved instances if need be (though most service providers may not like that since that costs $$)

-S




Confidentiality Notice: This e-mail message (including any attached or
embedded documents) is intended for the exclusive and confidential use of the
individual or entity to which this message is addressed, and unless otherwise
expressly indicated, is confidential and privileged information of Rackspace.
Any dissemination, distribution or copying of the enclosed material is prohibited.
If you receive this transmission in error, please notify us immediately by e-mail
at abuse@xxxxxxxxxxxxx, and delete the original message.
Your cooperation is appreciated.




References