RE: Methods in the NIM requirements

[remoore@us.ibm.com]

Walter,

There's been some discussion back and forth on this point, but I'm
not sure the key point has been made.  You said:

> If the NIM model implies a "go" semantic, how is the model applicable
> to the directory. If I have a method for reboot that means do it now,
> how do I represent the current state (rebooting/(re)initializing/
> unavailable) of the machine  in a directory.
> "Go" implies real-time semantics, yet we all know that directories
> are not well suited for real-time status representations.

Your unstated assumption here is that if the "Go" semantics are not
representable in a directory, then they cannot be a part of the NIM
model.  If we generalize this argument, we're forced to conclude that
if feature B is not representable in a database, then it can't be a
part of the NIM model either.  And if feature C is not representable
in an SNMP management system, then it can't be a part of the NIM
model.  And so on.

Unless you're willing to have some aspects of the NIM model be
unmappable to certain implementation technologies (as we've
acknowledged all along with CIM / LDAP:  we've always said that
CIM statistical information is not appropriate for a directory), then
of necessity the NIM model can be nothing more than the greatest
common multiple (in English, perversely, the phrase normally used
to describe this concept is "least common denominator") of all the
implementation technologies.  Do you really want to limit NIM in
this way?

Regards,
Bob

Bob Moore
IBM Networking Software
+1-919-254-4436
remoore@us.ibm.com