Re: Databases, was Data Processing Switch

[Tom Droege <tdroege2@earthlink.net>]

I just made 2 copies of the files to send to Michael S. (for the data base) 
and Michael R.  (To look at the data quality).  It is a five disk set so I 
don't really want to make a lot of copies.  Still if any of you data base 
fans want a set I will make one for you.  There are only about 3 full 
disks  15 million measurement pairs or so.  It is convenient to separate 
the older and newer data and by TOM.

Tom Droege

At 10:09 PM 6/2/03 -0600, Robert Creager wrote:
>On Mon, 2 Jun 2003 16:56:25 -0700 (PDT)
>Chris Albertson <chrisalbertson90278@yahoo.com> said something like:
>
> >
> > Rob,
> >
> > So, what kind of performance do you see, How long to import
> > 1E7 observations
>
>Depends.  If importing all measurements (i.e. all stars, not those just 
>seen both bands), I get from 6000 to 9000 per second (using COPY).  I've 
>just recently added a step to only import stars which match in both bands, 
>and that slows me down to about 1 image set per 10 to 30 seconds (around 
>100 per second tops), but I've not yet investigated ways to improve that 
>dB performance (I believe I can).
>
>I've just started trying to trim what goes in, as the UPDATE I run to 
>calculate the magnitude from color, zero point and