Subject: | |
From: | |
Reply To: | |
Date: | Thu, 19 Nov 1998 14:08:51 GMT |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
On Wed, 18 Nov 1998 12:53:58 -0500, Lee Gunter <[log in to unmask]>
wrote:
>I am generally against tier-based software pricing, in principle; however,
>I have seen more support work required of some of our ISV's caused by the
>problems apparently unique to running on multi-processor systems vs.
>single-CPU machines. This is merely an observation -- not a justification
>in and of itself for tiered pricing schemes..
>
Without any more specifics, I would say multi-processor problems are a
result of problems at the OS level, not the application level. No sane
application (unless it's a hardware monitoring tool, for instance)
should care whether the system has one or more CPU's, or if it
switches CPU's mid-process. If your payroll application (for instance)
*is* sensitive to that, then I would say the application is poorly
written, and you have created your own support headaches for which the
users shouldn't have to pay.
So I would think that multi-CPU support would in *no way* justify
extra cost. Again, it's the OS programmers who have to worry about
such things.
---
Mark Landin "For anyone who was never good at
T. D. Williamson, Inc. anything, technology has been a
UNIX Sys. Admin real boon" --- my mom
|
|
|