[pgrouting-users] Newbie: point_to_id and very large tiger_2010 tables

[Michael Papet]

Hi,

I'm a GIS newbie that has TIGER_2010 data for U.S.-California loaded in PostGIS as a pilot.

For California alone, the edges table stores the multilinestring data has 3 million rows.  I need to load/use all 50 States when I am done. Attempts to run assign_vertex_id on California alone go on for a few days then just dies with nothing inserted.

It seems to me that I should just add a column to the edges table (or a view or something related to edges) to permanently store the value returned by the point_to_id function instead of trying to generate a giant temporary table.  

Am I headed in the right direction?  What are the consequences I have to work through?  I am competent at editing the SQL functions.  I just don't know what the consequences of  storing the data generated in point_to_id function somewhere else has on other pg_routing functions. 

Thanks in advance for your patience answering a newbie question.

Michael.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/pgrouting-users/attachments/20110909/a6e32004/attachment.html