[Babase] MEMBERS/CENSUS/DEMOG conversion is running

Stephanie Combes babase@www.eco.princeton.edu
Fri, 24 Sep 2004 15:05:45 -0400


Hi Karl et al.,

 Very exciting.  I've set it to run myself here at Duke.  So far we're at 
about 24hrs and still going.  I think we might need some speed-ups.

  Karl - were you planning on coming back here to Duke for a bit in October?

-s

--On Wednesday, September 22, 2004 11:20 PM -0500 "Karl O. Pinc" 
<kop@meme.com> wrote:

> Hi,
>
> I've updated the conversion program to do MEMBERS/CENSUS/DEMOG
> and am running it now.  I'm using the babase_test database.
> Steph can run it for herself if she likes in the
> babase_sandbox database as usual.
>
> I believe I've finally got all the
> bugs out of the interpolation program.   However, performance
> is terrible with the 'old-style' census rows.
> It works just as it does now, and in normal use
> it does not matter as there are no new 'old-style'
> census rows, but the conversion is slow.
> This hot minute I'm going to move on and work toward
> getting the rest of the tables converted as they should
> go quickly.  The program may still be running the next
> time I look in a day or so in which case I can speed up
> the interpolation without much effort.
>
> On another note, the REPSTATS/CYCSTATS updating is
> terribly inefficent.  This will slow down the census
> updating as entering census data updates Statdate
> which updates REPSTATS/CYCSTATS.  I can make this
> much faster too with not very much effort and will
> probably need to do this before we go live.
>
> Karl <kop@meme.com>
> Free Software:  "You don't pay back, you pay forward."
>                    -- Robert A. Heinlein
>
> _______________________________________________
> Babase mailing list
> Babase@www.eco.princeton.edu
> http://www.eco.princeton.edu/mailman/listinfo/babase