>>I'm going to try to simulate something along the lines of what you have. I'll let you know if I find anything.
>
>Thanks, I believe this situation might be caused to the log file becoming too big. I do not believe there are maintenance on that one. So, it might have grown very big since the last two years and this type of operation could then take additional resources, lock or related.

I just did a test of simultaneous inserts and deletions at the same time, inserting rows in blocks of 100,000 at a time, and then inserting new rows while deleting old ones. Even thought it got progressively got a few seconds slower once I got to a very large # of rows, I never had a problem

I would look at the size of your log and the % of log space being used. You can get this for all your databases by running this command....

DBCC SQLPERF ( LOGSPACE )

If there's no maintenance on the log, it wouldn't surprise me that you might be having issues.