Compressing databases after mass deletions

Question

After deleting a bunch of flows, flow property factors, and impact factors using SQL and then manually deleting all "Locations" data, when I export a database it's still reflecting the size before all the deletions. Running compress database shows that the database will be the same size before and after. Is there any other way to get the database to meaningfully compress?

The work-around right now is to export all data to json-ld, and then import into a blank database. That database reflects the expected size.

Answer 1

Hi, the flows etc. do not need a lot of space, thus the database will not be considerably smaller after the deletes; openLCA uses here a function from the Derby database that is integrated in openLCA, and that may not run at all if it is not worth it (so to say). System processes and images are typically using a lot of space, deleting these typically shows an effect.

Comments

With the above, most of the data was in the Locations data - all the geo-jsons. I had a database shrink from 50MB to <1MB after removing all the locations (which weren't being used) and following the process above. Motivation was to be able to email the database to a client/sponsor.