We experienced a brief, 2-hour period of the site not responding to requests today.  The reason for it was a setting at the database level having to do with the percentage by which to grow the log file when required.  It was set at the default value of 10%, which was way too high for a database the size of GP's.  As requests continued to come in, the autogrow operation would time out, only to attempt to do it again a few seconds later.

The settings for the log file and the data file have been reduced to 1% and 2%, and the autogrow has already taken place without restart of any services.  I will continue monitoring everything throughout the day, just to be sure, however.