Open Stack

So that's part of scheduling; but likely only a part of it; it depends 
on the matrix of resources you want to schedule on (just compute node 
targets, or compute nodes + network resources or ...)

If u want, looking at how the filter scheduler works in nova is a good 
start.

https://github.com/openstack/nova/blob/master/nova/scheduler/filter_scheduler.py 
(prob. also read the new stuff around placement and claims and ...)

There has been other attempts at using constraint systems for this:

https://github.com/openstack/nova-solver-scheduler

https://review.openstack.org/#/c/210549/ was something I started a while 
ago (read for reference if you want).

IMHO there is a lot of research potential and experiments that can be 
ran, especially if you start to decrease the delay between 'actual 
resource set available' (what is really being used) and 'predicted 
resource set available' (ie what the scheduler knows about) [note there 
there is a eventual consistency problem here].

I'm sure there are bunches of other papers available to, including 
various ones from google borg and ...

So ya, imho a pretty 'fresh' area for things to be messed with...

Felix Fischer wrote:
> Am 2017-12-01 01:51, schrieb Joshua Harlow:
>> Felix Fischer wrote:
>>> Hey Folks,
>>>
>>> I'm Felix and I'm looking for a topic for my master's thesis in computer
>>> science. I studied at the Humboldt University of Berlin and currently
>>> I'm working as an OpenStack consultant. I was hoping some of you could
>>> point me to areas or even specific topics related to OpenStack which
>>> need further scientific research. My topic needs to be about a
>>> scientific research question and cannot be a programming task or
>>> something similar to that. Maybe some of you have a problem for me which
>>> needs solving or can point me directions where to look at or who to ask?
>>>
>>
>> Do you like scheduling problems (research and practical), because
>> there are a lot of them? :)
>>
>
> By scheduling problems you mean "decide on which compute node to create
> a new instance"
> then yes :-) That was the direction my co-workers pointed me to, as well.
> Do you have any further information or papers to read?
>
>>> Thank you in advance and kind regards,
>>>
>>> Felix Fischer
>>>
>>> _______________________________________________
>>> Mailing list:
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>> Post to : openstack at lists.openstack.org
>>> Unsubscribe :
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack