Hi all.
I am currently working on a data-import module that takes up to 1000 XML records and dumps them into a MS Access table. During this process, the data from the Access table is loaded (the system uses an OR Mapper for the Access-.NET mappings) and compared against the new XML data to find any duplicates (this can not be done in a query). The problem is that, even though I am using paging to go through the existing records at 250 a time, the memory consumption of the application triples (up to 70mb) and does not come back down. If another import is run during the lifetime of the application thread, the memory consumption soars to 120mb. This continues for every import that is run.
How can I stop this The mapper does not leave any resources open. I have tried setting all data collections and potentially large objects to Nothing. I have even let the program stand idle for half an hour, but the memory usage just will not come down.
Regards.
Stephen.

Controlling memory consumption...
R Parry
William Klein
Well, we first tried it with our in-house mapper, but it has a large overhead when processing relationships, so we're now using the Wilson ORMapper, but neither seem to have any diminishing effect on the memory usage. FYI, we are using custom, strongly typed, collections (Inherited from ArrayList).
Thanks again.
Stephen.