On 11/14/10 11:43 PM, Bjoern A. Zeeb wrote:
On Sun, 14 Nov 2010, Julian Elischer wrote:
Julian,
this sort of dovetails into something I've been thinking about for
a while,
which is NUMA support.
really good thoughts, but bad timing and wrong target audience I
think.
I don't think the thought was seep enough to warrant a larger
audience :-)
1) first of all you want the basic underlying OS support all sort of
things and as we know people are working on things here and there
and it's not only NUMA.
Well yes, of course but it's interesting how various different parts
of what are until now
completely separate projects might some day come together in a larger
picture.
2) before 1) is rock solid I will not even think about how to leverage
it for VIMAGE. It would really just be a waste of time; it will
however be a good idea to make sure that the proposals people come
up with in 1) will scale for VIMAGE as well, but that's a different
story. I am at least trying to keep an eye on the network stack
side.
3) Before VIMAGE is NOT stable we MUST NOT introduce more complexity.
agreed. My only thought at the moment is to try and keep an eye on the
bigger picture when changing the smaller parts.
So if you are eager to work on this I am all for it and you should get
in touch with the folks doing 1).
/bz
_______________________________________________
freebsd-virtualization@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization
To unsubscribe, send any mail to
"freebsd-virtualization-unsubscr...@freebsd.org"