Jet Profiler does not collect all traffic to the database server, it
only polls every x milliseconds. So it sees all queries that are
executing at the time of polling. But any queries that are executed between
two samples are missed. Therefore Jet Profiler cannot tell how many
times the query has run during a time frame.

For this reason we have the thread load measure, where we try to
visualize the observed frequency, expressed by how many threads are
observed running the query at every sample time. A value of 1 means the
query is always running (by one thread on average).