<html>
<body>
<font size=3>At 04:31 PM 9/28/2004, Roland Dreier wrote:<br>
<blockquote type=cite class=cite cite=""> Michael>
It was not specified due to the lack of standards for<br>
Michael> higher-level service management which
partitioning is<br>
Michael> classified. Given there is an OpenSM
effort in flight<br>
Michael> and the Service ID spec is already in
existence, it isn't<br>
Michael> that tough to acquire an IP service ID (or
any other<br>
Michael> protocol that one wants to support) and
implement a<br>
Michael> solution along the lines that I describe
above. This<br>
Michael> would lead to a more dynamic environment
while reducing<br>
Michael> the impact to the administrator.<br><br>
This scheme might indeed be reasonable. However, given the absence
of<br>
an IBTA spec or IETF draft, I don't see how we can rely on it in
our<br>
IPoIB driver right now.</blockquote><br>
The IBTA defined the specifications to establish standard wire protocols
to discover services without having to track all potential
services. The IETF does not address how P_Keys are managed only
that the IP over IB component must follow a set of semantics / operations
to enable IP communication; all IB-specific management issues are outside
the scope of the drafts. <br><br>
As another person put it - embrace and extend. The approach being
advocated is something that will be interoperable and provides a solution
that can be easily incorporated into all IP over IB
implementations. It also clarifies the ambiguous IB - admin
management interactions for IP communications.<br><br>
Mike</font></body>
</html>