At 04:57 PM 10/14/2004, Fab Tillier wrote:
> From: Michael Krause [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, October 14, 2004 4:21 PM
>
> Why not just leverage the SDP port mapper protocol already defined in the
> RDMAC version and avoid having to provide a new address family?  The port
> mapper protocol is interconnect independent and will enable sockets
> applications to more easily be executed transparently.  It seems
> counterproductive to continue to pursue a new address family.
>
> BTW, the new port mapper protocol will also work with the new async
> sockets and memory management API that is nearly complete (should be
> approved soon within the OpenGroup).  This would greatly enhance socket
> application design and provide greater performance when operating over a
> RDMA interconnect than traditional BSD sockets.
>

A while ago, there was some discussion of having transparent port mapping be
a bad thing, and a security vulnerability of some sort.  Note that I don't
personally believe that.

The SDP spec submitted to the IETF as a draft has the port mapping.  The HP IETF gurus did not have a problem with the port mapping.  Discussions with various IETF AD did not generate a negative reaction but one never knows how the IETF will act.  In general, the exchange has measures to mitigate any DOS attacks and conserve resources.  Take a look and please consider as this is the best way to get a large number of sockets applications to quickly operate over RDMA interconnects without requiring any source code changes. 

Mike
_______________________________________________
openib-general mailing list
[EMAIL PROTECTED]
http://openib.org/mailman/listinfo/openib-general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Reply via email to