To ascertain the overhead values and
, we must
measure the instrumentation code that calculates inclusive and
exclusive performance. However, the
and
values
may be at the same scale as the measurement precision. Thus, we must
construct a test that guarantees statistical accuracy in our
estimation of
and
.
Our approach is to conduct a two-level experiment where the inclusive
measurement code is executed number of times (in a tight loop) and
the performance data being profiled is measured and stored. This
procedure is then repeated
times and the minimum performance value
across the
experiments is retained. We use that value divided by
to compute
. While this does not entirely insure
against anomalous timing artifacts, its statistical safety can be
improved by increasing
. The two-level experiment is then
repeated, this time with the exclusive measurement code included,
resulting in an approximation of
. We can
substract our
approximation to find
. It is
important to note that we need to do this overhead evaluation at the
beginning of the application execution for the specific set of profile
performance data being measured.