Capture a timestamp upon call to BEGIN_TASK and at the end of END TASK to track the intrusion of ILO
Write this information to the ILO_RUN table in 2 new columns (ILO_BEGIN and ILO_END).
It may be necessary to rename the current BEGIN and END dates to TASK_BEGIN and TASK_END;
from Doug Gault:
I'm unclear as to how this will actually capture the measurement intrusion.
If a timestamp is captured at the start of BEGIN_TASK and at the end of
END_TASK, how would you extrapolate the time actually spent within ILO from
those two timestamps.
I would think that it would be better if ILO were programmed to capture
the duration spent within itself at each call (Begin and end) and write
that duration (or two, begin and end duration) to the ilo_run table.
Also, if you're capturing the end timestamp before data is written to the
table, then you're missing part of what you're supposed to be capturing in
the intrusion factor.
I think that this can be done, but I don't think that it will be 100%
accurate and should be documented in the "caveats" section of the