The local water and electric board saw it fit to shut off everyone in my neighborhood’s electricity for 4 hours this morning. They chose the incredibly convenient time of 8 am – noon to do it. For the early risers this isn’t an issue, but for normal human beings like myself this is a problem. I had to take a cold shower, and then dress in the dark. And now that I’m looking at myself, it’s pretty obvious I dressed in the dark.
But there is a bigger issue at hand. I almost logged in to our lab’s server from home to start the next series of calculations last night. Almost. It was actually one of my finer moments, I was contemplating leaving early (8 pm) and just starting them from home, when I decided to just suck it up and get them going before I left. It’s a good thing, too. No power at home = computer off = remote login session killed = no calculations.
I have been traumatized by power outages before. This summer the university experienced an unexpected blackout due to failure of the cooling system at facilities services. Again, no power = computer off. I just happed to be about a month into a 3-4 month calculation. We received emails saying that the outage was a fluke and it was safe now, it would never happen again. Whatever. I restarted the calculations. A week and a half later I come in to find the server off again. ” WTF,” I said in the unabbreviated form. Another outage, same reason. The Patron Saint of Graduate Students was not looking after me this summer. And then two weeks later we had an annual campus wide electrical shutdown (a rarity among universities). So those two weeks were useless. This wouldn’t have been a problem had the original author of the program (not me) written in some periodic safeguards against unexpected terminations. But this is like kamikaze code, all or nothing. I guess I could do it. But I hate coding. Also, It wouldn’t be so bad if they didn’t take 4 months to run to completion. It’s got to be the most inefficient code ever written.
Power outages at a research institution can be devastating. The series we suffered could have had catasrophic effects. Lasers down, servers down, STM’s and TEM’s down. Not to mention particle accelerators. The time it takes to get everything up and running again is a phenomenal waste of person-hours. It put me back at least 6 months. And I’ve got easily resumable research. It takes a long time to catch up to where I was when things stopped, but it’s almost effortless on my part.
Now I am deathly afraid of power outages and am scared to start more calculations, only to have them terminated resulting in my premature stroke.