RE: SQL vs. traditional I/O?

comments inline...

> -----Original Message-----
> From: midrange-l-bounces@xxxxxxxxxxxx / Pat Barber
> Sent: Thursday, July 22, 2004 4:06 PM
>
> performance issues continue to nag the whole thing.

Is this still true?  I thought well-designed SQL will perform as well as, if
not better than, well-designed native I/O, for most applications as of one
of the V5 releases.

> Any time somebody tells you SQL is god's gift to
> programming has been to too many IBM seminars.

<g>

> I recall several stories by Jim Sloan where he compared
> opnqryf(building views) to the lowly fmtdta, which in some
> cases would beat the pants off opnqryf(sql).
> That's plain old "Sort & List".

Agreed, especially with the qualifier "some cases".

> In fairness to SQL, it can do record selection and certain
> other functions quite well, but in the wrong hands, SQL is
> a system killer.

In the wrong hands, native I/O can be a system killer.

db