News from stress central
Sep. 3rd, 2008 10:10 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
sawylThe last couple of days have been somewhat stressful. Not only have I been frantically working on my last big philosophy essay of the year, but I've also been preparing to go away for a few days and getting my respectable clothes dry cleaned.
As if all that wasn't enough, work hasn't been great either. One of the scientists spotted signs of subtle data corruption in their output, kicking off a full on witch hunt — there's nothing that frightens computational climatologists and meteorologists quite as much as the idea that all their output data might be every so slightly wrong, silently propagating bad results through their models. Consequently, everyone was on edge and much pressure was applied to locate the cause of the problem.
After much winnowing, we were eventually realised that the initial thought that the problem had been occurring system-wide was incorrect and we were able to focus our search on one particular component. We eventually traced the problem back to a fibre channel connection that was inducing single bit errors in the output data stream and we were able to pass all the problem over to the hardware guys to sort out.
With work sorted, I then rushed home to get myself packed and to the station in time for my train. The journey was actually pretty good: the carriage was quiet and relatively empty; I had a table seat; I was able to get a whole load of useful philosophy done; and it wasn't even raining when I arrived.
As if all that wasn't enough, work hasn't been great either. One of the scientists spotted signs of subtle data corruption in their output, kicking off a full on witch hunt — there's nothing that frightens computational climatologists and meteorologists quite as much as the idea that all their output data might be every so slightly wrong, silently propagating bad results through their models. Consequently, everyone was on edge and much pressure was applied to locate the cause of the problem.
After much winnowing, we were eventually realised that the initial thought that the problem had been occurring system-wide was incorrect and we were able to focus our search on one particular component. We eventually traced the problem back to a fibre channel connection that was inducing single bit errors in the output data stream and we were able to pass all the problem over to the hardware guys to sort out.
With work sorted, I then rushed home to get myself packed and to the station in time for my train. The journey was actually pretty good: the carriage was quiet and relatively empty; I had a table seat; I was able to get a whole load of useful philosophy done; and it wasn't even raining when I arrived.
