[mapguide-users] Database Choice

[Zac Spitzer]

Performance wise i'd stick with SHP or SDF for static data and only
use a db when you need to for dynamic layers.

Generally speaking in terms of database design, you need to think like
you are the database and mapguide engines.

Mapguide is going to run a separate database query for each layer in a
map, so putting all of your polygons in a single table means the
database will need to sort through all of your polygons,  just to
display a subset of the table.

Repeat that for each layer using polygons in your map. it's starting
to sound like a lot of work.

Plus updating your data is going to involve updating some very large indexes.



Zac

On 5/15/07, Andre Schoonbee <andresch at iway.na> wrote:
> Dear List
>
> I hope I am at the right door for this question.
>
> We are busy planning a Mapguide deployment which will eventually cover the
> entire country's special data as can be found at our Local Surveyor Generals
> office. This will also include all National statistics data etc. So we are
> looking at a few 100 GB.
>
> Our first step will be to start small and then grow to include all later.
>
> My questions are:
> 1) We plan to have all data on PostgreSql, or would oracle be a better
> choice. The reason why we prefer Postgresql is because is't free.
>
> 2) Database design: What would be the best design concept? We will start
> with hundreds of *.shp files. Do we create 1 database and only a few tables
> such as polygons, polylines, points; or should we differentiate further? I
> really need some info on best practice and for optimal performance.
>
> PLEASE, any guidance would be appreciated
>
> Andre
>
> _______________________________________________
> mapguide-users mailing list
> mapguide-users at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/mapguide-users
>


-- 
Zac Spitzer
+61 405 847 168