[postgis] Big Database
Timothy H. Keitt
tklistaddr at keittlab.bio.sunysb.edu
Wed Oct 10 09:20:58 PDT 2001
Try this link:
I converted the full resolution global coastline database from GMT
(http://gmt.soest.hawaii.edu/) into postgis format and dumped the
database into the above file. Its 43MB compressed, so it should be a
good test data set. Unfortunately, postgresql puts the database name in
the ascii dump file, so you might need to edit the top of the file and
replace "tkeitt" with your database name. (Since its a big file you
might use 'head' and 'tail' to cleave the first x lines from the top and
'cat' the result together after editting.)
Roderick A. Anderson wrote:
>Several weeks ago I started (again) asking questions about coordinate
>systems and PostGIS. The plan was to test the feasibility of a Regional
>level database - the Inland Northwest (north Idaho, east Washington,
>west Montana, north Wyoming.) I have about 170 USGS quads of
>Cartgraphic Feature File data (CFF) and the same amount of USGS DEMs
>all in UTM zone 11N.
> The DEMs proved to be too much. I ran out of disk space on my test
>system at about 7 million points and besides how do you test PostGIS
>when all you're querying is points.
>The CFF data was more interesting but only worked out to about 200K
>tuples of line data. So I went looking for more data. Found a bunch of
>free stuff but was disappointed to see it was in LAT/LONG making it less
>than optimal for combining with my zone 11N stuff. 8-(
> I'm now at the point of figuring out how to get the two of them into
>one 'coordinate' system so I can crank up the tuple count up.
>While doing this I think a found a problem with PostGIS - or actually
>OpenGIS. If I understand it correctly you can only have one data type;
>point, line, polygon, in one coordinate system; UTM 11N in a table.
> This seems to go along with the way most GIS software wants to handle
>the data but is kind of limiting from a database perspective. If I want
>all the data in an area (BOX) why do I have to query several tables? And
>shouldn't my GIS or COGO software transform the data to whatever system
>it (or I) want it in. I see a way around some of these limitations but
>it would be a kludge (use UNIONs) and could fail on a REALLY big query.
>So I'm still looking and testing but wanted to let those that have been
>so helpful know that I hadn't stopped working on this.
Timothy H. Keitt
Department of Ecology and Evolution
State University of New York at Stony Brook
Stony Brook, New York 11794 USA
Phone: 631-632-1101, FAX: 631-632-7626
To unsubscribe from this group, send an email to:
postgis-unsubscribe at yahoogroups.com
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
More information about the postgis-users