On Mon, Jan 10, 2011 at 04:33:30PM -0500, Jay Ashworth wrote:

----- Original Message -----

...

From: "Valdis Kletnieks" <Valdis.Kletnieks@vt.edu>

...

...

Why the hostility, Valdis?

As I said several times - it's not hard to be 98% or 99% sure you can make all your commitments. However, since predicting the future is an inexact science, it's really hard to provide a *100% guarantee* that you'll have enough contended capacity to make all the performance targets even if every single occasional customer shows up at once. As Jay pointed out in his follow-up note, his backup strategy is "scramble around and hope another provider can come through in time", which is OK if you *know* that's your strategy and are OK on it. However, blindly going along with "my usual provider guaranteed 100% availability" is a bad idea.

I don't think Kelly is on his first rodeo, and I know I'm not.

"scramble around" is a bit pejorative as descriptions for my booking strategy go, but everyone has a cranky day every so often, not least me.

:-)

And note that I *also* pointed out that carrier statmuxing on the transport is a valid strategy for capacity elasticity, in that particular environment.

...

Remember, we're coming out of a solar minimum. ;)

Are we in fact coming out of it yet? I heard it was getting deeper, and that we were looking at a Dalton, if not another Maunder.

I'll have to find the paper I read yesterday that said we should expect to wait a long time before we see sunspot counts back where they should be. ... Try this: <http://news.sciencemag.org/sciencenow/2010/09/say-goodbye-to-sunspots.html> -- Mike Andrews, W5EGO mikea@mikea.ath.cx Tired old sysadmin