Today, one of the server I/we have applications on, we have experienced timeout issues at two occasions. On both of them, only the transactions (Robot, Web Service, Web applications) that were trying to update on two big tables were involved. Thus, all big SQLs involving those tables didn't cause any timeouts. This is an environment which is stable and nothing in the last two weeks could have been changed to create such an impact.
Network analysis and such are searching for the cause of the problem. I was wondering if the UPDATE command on SQL Server requires more CPU time than pretty much everything else. As, on all those errors, there were only related to updates on one of the two tables. It seems as if the update functionality on those two tables became lock for a few seconds. This pattern was identical in an elapse time of two hours. During that period, on either one of them, before and after everything was running as fast as usual.