As we grow, we’ve been building out larger grids for our demonstrations lately. In fact, during the recent ISPcon show in Santa Clara I was using three grids for my demos, the largest of which was 32 servers provided by our partners at Layered Technologies.

Well, as is innevitable when you’re running a large number of servers, I’ve had a few give up the ghost lately during demos.
Including one during the show. After all, they’re just tin and silicon.

Servers dying during demos would ordinarily be the stuff of nightmares. I’ve lived through those nightmares many times over the years while introducing dozens of new products. My song and dance skills have garnered me job offers more than once. Not this time!

Instead, these server failures simply caused AppLogic to do what it does best - separate software from the hardware that runs it. When the servers died, AppLogic simply restarted affected apps on other resources as the customers and I watched.

The geek in me wants to start a collection of flaky servers to build a grid from just so I can see how far we can take the process. The stranger the failure the better. Any donations?
!!![4] Comments!!!