![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
How Not to Design a Pocket Program
May. 12th, 2006 01:50 pmI am sitting here here watching my secondary (previously the primary, now demoted) computer spin its wheels as I try to do an objective test of how much faster the new machine is. I fed the new machine one of my optimization models and it took 20 minutes to digest it. I'm running the same model on the secondary machine. It has been nearly three hours so far. I fear that it will still not be finished by the time I need to leave to head up to The City. So I have at least bracketed the run time as being at least three and less than about eighteen hours, as I set the same job up yesterday before leaving, and it was finished when I came in this morning.
So, not having much I could get finished, I went through my WHC membership package, sorting the things I need to keep with me (not much) from those I can leave behind. Among the things I need is the Pocket Program. I looked through it to see if there was anything I wanted to attend.
( Whose Bright Idea Was This? )
Good thing I didn't really expect to attend much programming, anyway.
Aha, that job just finished, taking just over 3 hours (180 minutes) versus 20 minutes on the new machine. So that's a ninefold improvement in model run time. I am pleased.
So, not having much I could get finished, I went through my WHC membership package, sorting the things I need to keep with me (not much) from those I can leave behind. Among the things I need is the Pocket Program. I looked through it to see if there was anything I wanted to attend.
( Whose Bright Idea Was This? )
Good thing I didn't really expect to attend much programming, anyway.
Aha, that job just finished, taking just over 3 hours (180 minutes) versus 20 minutes on the new machine. So that's a ninefold improvement in model run time. I am pleased.
