RE: What About Price vs. Performance?

I did EXACTLY that, Nathan, two or three weeks ago.  I think I did 200
pageup/pagedowns.  I got the following numbers:

"A well written web application can serve up a 3KB dynamically generated web
page in about 100msec [of CPU time], as opposed to a green screen
application outputting 1920 characters (or less) in about 20msec."

About a 3-1 KB/sec ratio (or less).  Web was 30KB/sec, while green screen is
a little less than 100KB/sec (or maybe a lot less, because I doubt that it
painted the entire screen on the pageup and pagedown).  Of course, I'll have
to do far more stringent testing than this to reduce the statistical error,
but a 3-1 ratio isn't bad.

Joe


> -----Original Message-----
> From: owner-midrange-l@midrange.com
> [mailto:owner-midrange-l@midrange.com]On Behalf Of Nathan M. Andelin
> Sent: Wednesday, April 04, 2001 5:25 PM
> To: MIDRANGE-L@midrange.com
> Subject: Re: What About Price vs. Performance?
>
>
> >I'll be doing some CPU timings, but subsecond ain't bad.
>
> Joe,
>
> With your revitalization work, you're probably in a great position to draw
> some comparisons between 5250 and Web applications.  You might try hitting
> the PageUp key on the 5250 screen about 100 times and track the
> elapsed CPU
> time.  Then do whatever corresponds to PageUp in the browser 100 times and
> track the CPU time.  My guess is that the difference is vast.
>
> I wonder if IBM and it's customer really know how much more CPU time a Web
> application takes.  It might make a difference in pricing if this kind of
> data were published.
>
> BTW, your TCO points are excellent.  Thanks.
>
> Nathan.

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---