No need to apologize but that is no what I was refering to. Given that
almost everyone, these days, is writing to RAM caches (Apple, EMC, LSI,
NetApp, etc.) that wasn't my concern.
I believe it was Howard Rogers who, not necessarily first but often and
well, discussed this here at c.d.o.server. But then I've been wrong
before and may well be wrong again.
Daniel A. Morgan
(replace x with u to respond)