[Orca-users] Graph scaling

[Ziegenhorn, Eric]

I was wondering if there is a way to control the scaling of a particular
graph in Orca.  I'm considering two situations.  One is that outliers screw
up the scaling of a grpah.  For instance, 99% of the data points are between
0.0 and 1.0, but one value is 64.3 thus making the graph scale from 0 to 100
and making the important data between 0 and 1.0 impossible to see.

The second situation is one I have an example of in front of me.  I have CPU
Usage graphs from two servers.  On both, the graph is scaled from 0 to 100
percent.  In the detailed numbers below the graph though they are differnt.
Server #1's measurements run from a minimum of 357.778m to a max of
22269.444m, excluding Idle measurements.  Server #2's measurements run from
a minimum of 0.347 to a max of 12.061, also excluding Idle measurements.
So, in table form:

	Server1	Server2
	-------	-------
Min:	0.357		0.347
Max:	22.269	12.061

Here I have converted 357.778m to 0.357, correct me if I am wrong.  What I'm
wondering is why the two graphs aren't both in "m" units, or just regular
units.  Server1 has the greater range of values, but has values with the "m"
suffix.  The difference makes the values hard to compare against one
another.

So, can somone explain the difference between these two graphs?  And, does
anyone know how to force the scaling of a graph to a particular range of
values and/or a particular unit of measurement("m", "M", etc)?

Any help would be greatly appreciated.  Thanks, Eric