[OSM-dev] Ideas for speeding up the TIGER import

Martijn van Oosterhout kleptog at gmail.com
Sat Sep 1 17:52:16 BST 2007


On 9/1/07, Jon Burgess <jburgess777 at googlemail.com> wrote:
> I think our present Tiger import process is fine for medium sized data
> sets, just not for Tiger sized data sets. If we take the AND data as a
> more typical example, it has ~10M objects. The tiger import is
> progressing at ~30 objects/s which means the AND import would complete
> in around 4 days. This seems reasonable to me.

That's one thing I forgot to mention in the previous email: if the
TIGER dataset is anything like the AND one, then the change to remove
segments will cut the data volume by 40%. That is a *big* saving.

I do agree the disk usage issue is troubling....

Have a nice day,
-- 
Martijn van Oosterhout <kleptog at gmail.com> http://svana.org/kleptog/




More information about the dev mailing list