not really.
the idea is that to an amazingly great extent, you should be able to
automate b) completely.
and in a), the two models would mesh already (because there is one
ubermodel a la CIM).
On Nov 27, 2007, at 4:16 AM, Thomas Delaet wrote:
In my opinion, option b) is an extension of option a).
Option a) assumes a centralised model control body. In practice this
implies that if I develop a model for managing webservers and someone
else defines a model for manging DNS servers that I would like to use
in my infrastructure, I'll have to do a lot of manual work to
integrate those two into a consistent static model.
Option b) has the potential to automate (at least parts) of this
process. In this case, you still have to manually map relationships
between the two models, but they don't have to be adapted ...
Does this makes sense?
Kind Regards
Thomas
On 11/20/07, Andrew Hume <[EMAIL PROTECTED]> wrote:
i don't think so. although i was somewhat unaware of this
terminology,
this is not what i was getting at.
to be clearer, the choice is between a (relatively) static
globally shared
information model,
and a much more dynamic situation where a new service may bring a
largely
incompatible
ontology with it, but if the part needed for initialisation and
activation
does mesh
with the existing services, then we're okay.
On Nov 20, 2007, at 1:11 AM, Steven Jenkins wrote:
On Nov 19, 2007 12:35 PM, Andrew Hume <[EMAIL PROTECTED]>
wrote:
...
it seems like there are just two answers:
a) there is a global model, populated by things like port numbers,
main
memory,
cpu load, available disk space, firewalls and so on. all services
can state
their requirements
and measure their usage and performance in terms (predicates etc)
of these
entities.
b) the tool inherently knows nothing. it somehow discovers the
services
extant in teh cluster
and then figures out what to do by grubbing around through teh
ontologies
for each service.
so when the web service says it needs a port number as part of its
installation, the tool
finds the entity 'port number' as part of teh models belonging to the
'firewall' service
and the 'tcp/ip stack' service for a node.
You seem to be describing the debate between single and multiple
dispatch:
e.g.,http://en.wikipedia.org/wiki/Multiple_dispatch
That's not to say one or the other is better, but I think it would be
beneficial for people to take a look at the programming language
research on the subject.
Steven
------------------
Andrew Hume (best -> Telework) +1 732-886-1886
[EMAIL PROTECTED] (Work) +1 973-360-8651
AT&T Labs - Research; member of USENIX and LOPSA
_______________________________________________
lssconf-discuss mailing list
lssconf-discuss@inf.ed.ac.uk
http://lists.inf.ed.ac.uk/mailman/listinfo/lssconf-discuss
--
Thomas
_______________________________________________
lssconf-discuss mailing list
lssconf-discuss@inf.ed.ac.uk
http://lists.inf.ed.ac.uk/mailman/listinfo/lssconf-discuss
------------------
Andrew Hume (best -> Telework) +1 732-886-1886
[EMAIL PROTECTED] (Work) +1 973-360-8651
AT&T Labs - Research; member of USENIX and LOPSA
_______________________________________________
lssconf-discuss mailing list
lssconf-discuss@inf.ed.ac.uk
http://lists.inf.ed.ac.uk/mailman/listinfo/lssconf-discuss
