RE: SQL vs native access

> From: Jon Paris
> 
> With regard to Joe's comment on why the SQL guys found it frustrating.
> You
> have to remember that ever since SQL became available they have been
under
> huge pressure to improve performance because native I/O performance
used
> to
> make SQL access times look just plain silly.  The guys were constantly
> beaten up about it at every COMMON, on the internet boards, etc. etc.
The
> frustration came about because they wanted to go out an shout "look at
us
> we've improved performance 80%" - but they couldn't because the
changes
> they
> made had just improved the native IO by 50%!  They didn't get any
credit
> for
> that - just more complaints about being slower.  It would get
depressing
> after a while don't ya think?

Not if they just changed the damned playing field and said "we're going
to make SQL absolutely SCREAM for queries - if you need to make record
at a time updates, use native I/O."  And indeed, that seems to be
exactly what they've done.  My read of the redpiece is that the SQE was
designed by some really smart people who understand what SQL is supposed
to be used for.

I've always thought that the idea that SQL should somehow supplant
native I/O was a waste of resources.  I hope this new focus will end
that silliness.

Joe