Nick writes:

> Well brevity is a virtue but I hope you are better soon.

thanks!

> This Image thing is a puzzler:

Yep.

That's why I want to know what state processes are in most of the
time.  If 40% of CPU is used, what's happening the other 60%
of time?  Either waiting for disk I/O or for inter-process locking
of some kind.  Drastically different, and if user provides a little
more information, it probably eliminates one path or the other in
the analysis.  If they're reading from same flat file, could be
contention for the PLFD (Process Local File DEscriptor, where the
file's current record pointer is kept).

> 6.  Is access to the root file required for each put?  I think not

Shouldn't be for autodefer, I think.  Even if it were, it's in memory
as you point out.

> Is there a kind of bug in Image where the code uses the same semaphore
> even
> though different data bases are involved?

I hope not!

> I know.  I wonder what the test performance is like when only one
> data base is rebuilt at a time.

Good question.

--
Stan Sieler                                          [log in to unmask]
P.s.: please forgive typos/brevity,      http://www.allegro.com/sieler/
      I'm typing left-handed for awhile.