I'd go for SQLite or Valentina.
trying to make sense out of 1 million records, you need a good query
language, one that is able to do more than one "operation" with a
single query. Just imagine looping your indexes over and over again
trying to find the cross-references you're looking for. Using a
"real" database in this case will save you time and effort.
Andre
On May 30, 2007, at 3:55 PM, Mark Smith wrote:
On the other hand (and I'm not actually advocating it, since I've
never tried it), it would also be possible to build indexes of the
data that are kept in memory, while keeping the actual data in a
collection of many stack files which are then loaded and unloaded
as required.
As I say, I've never tried this (or needed to), but it might be an
interesting experiment sometime.
Best,
Mark
On 30 May 2007, at 19:48, J. Landman Gay wrote:
The original data I'm looking at importing is well and truly over
one million records.
I think that pretty much requires an external database. If you
dump that much data into a stack I don't think you'll like the
results.
_______________________________________________
use-revolution mailing list
use-revolution@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your
subscription preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution
_______________________________________________
use-revolution mailing list
use-revolution@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution