I had two different databases that I ran a cursor update against, which was executed from an SSIS job. Both were allegedly the same database as far as schema, size, record counts, etc, but they were backups from different points in time (one from 3+ months ago and one from about a month ago). When I ran the cursor update against the 3+ month ago db, it took 6 hours to process. When I ran the cursor update against the one month database it took 45 minutes. Close to the same number of records were cursor-ed and updated. Also both databases were processed on the same server. Any ideas what could have caused this increase in performance or what perhaps could have accounted for the cursor processing time? Also, the only thing running on the server was the SSIS job and the cursor update.