On Thu, 2004-06-17 at 12:58, Ranga Nathan wrote:
> Interesting. So it seems that if you want to track CPU and memory
> performance you have to do it at the aggregate VM level.

Or use Barton's ESALPS product, which is designed to do exactly that.

>  That makes it
> difficult to size the guests. How does one know how much resource to give
> to a guest. I do understand that if the guest is under-performing then you
> give it some more. But by how much?

My general rule of thumb is to start with small guests--128MB or so.
Bump up their storage as necessary until they are just-barely-into-swap
in normal operation.  Put your swap in VDISK, or at least your
high-priority swap.  That way when you *do* swap it isn't actually that
much of a performance hit.

> Very intersting questions...!

The thing NOT to do is start out with huge Linux guests.  A 512MB guest
will usually perform worse than a 128MB guest on a busy system, because
the bigger guest is always having to wait for VM to page his storage
back in.

Adam

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to