On Wednesday, 01/24/2007 at 09:46 CET, Rob van der Heij 
<[EMAIL PROTECTED]> wrote:
> While standing at the virtual white board... Since CP can already
> provide some form of virtual QDIO devices, it would seem more
> attractive to enhance that into virtual FCP. The good thing about that
> is also that it allows CP flexibility in reordering the I/O requests
> and perform them in a way that makes sense on a global level. Some
> interaction between CP paging and this virtual FCP might be very
> attractive.
> The scope of things like WWPN in Linux would be local to the z/VM
> system, and up to CP to interpret (e.g. to distinguish between swap
> space and real data).

This comes under the heading "Be careful what you ask for - you might get 
it."  :-)

At first glance that looks really attractive.  However, when you actually 
start wanting to manage your storage and performing end-to-end storage 
provisioning, you find that the question "Who is connected to what?" is 
difficult to answer correctly when there are two different views of the 
world.  WWPNs are really [fabric] network addresses and, like vMACs on 
layer 2 VSWITCHes, things work better and with fewer surprises when we 
avoid synthetic addresses.

That isn't to say we couldn't have vFCP with N-PIV WWPNs reserved for a 
particular user, but we already have that now, essentially, by dedicating 
the correct RDEV to the user.

Alan Altmark
z/VM Development
IBM Endicott

Reply via email to