|
|
|
|
|
|
|
|
|
|
xen-users
Re: [Xen-users] Basic Xen Questions
Ritesh Raj Sarraf wrote:
1) If I patch my kernel with Xen, will I be compromising on any of
the other
features that I've patched ? (Namely UML and swsusp. I use UML a lot)
swsusp: Software Suspend
http://sourceforge.net/projects/swsusp
It's unlikely this will work with Xen. This would undoubtedly
require a patch into the Xen hypervisor to save & restore the other
domains. Even if you were only running dom0, this might still not work.
UML: User Mode Linux
Interesting. UML is a technology which accomplishes some of the same
goals as Xen. Xen can run multiple instances of Linux, like UML. So
from that stand point, you'll get many of the same features as UML.
However, Xen can also run other operating systems: FreeBSD, NetBSD,
Minix (in theory), Windows XP (with the right processor), etc.
It's my opinion that Xen is a superset of the functionality one can
get with UML. If UML is important to you, I encourage you to
continue learning about the virtues of Xen.
The answer to the question: "will I be compromising on any of the
other features?" is yes.
2) So when I boot with a Xen kernel, the OS it boots into is dom0
which
simply is my normal OS. Right ?
More or less. However, the Linux running as dom0 will likely not be
the same -kernel- image that you are currently running. But, it will
be Linux and it will see your current file systems, devices (assuming
your configuration files are suitably identical), etc.
3) If I'm correct about Question 2 then, "when only using dom0,
does it add
any virtualization overhead to the host OS i.e. dom0 ? Or does it
simply
function as a normal Linux Kernel with Xen features not being in
use ? "
It does add overhead to use Xen even with only dom0 present, but
they're minimal. It's likely you won't notice them. Various
presentations indicate the performance impact is as low as 3%.
-- Randy
_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users
|
|
|
|
|