| Title: | DECWINDOWS 26-JAN-89 to 29-NOV-90 |
| Notice: | See 1639.0 for VMS V5.3 kit; 2043.0 for 5.4 IFT kit |
| Moderator: | STAR::VATNE |
| Created: | Mon Oct 30 1989 |
| Last Modified: | Mon Dec 31 1990 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 3726 |
| Total number of notes: | 19516 |
Does the DECwindows Program Office plan to publish the results of its
Characterization study??
We in the field desperately need curves showing use of resources for
typical applications. I'm thinking here of curves showing client and
server resources being used, (memory, cpu, net, mass storage perhaps),
for low imaging content up to high imaging content.
From forms like applications to highly interactive graphical
applications. And from Vaxstations, Decstations to VT1000's with mixes
of clients and server.
Please don't shrug your shoulders and answer me that my mileage may
vary. We all know that. But in order to help our customers configure
the most cost effective systems, we have to be able to guide them in
the right direction.
If we are looking at large customers, then cost means everything from
cables for networking to computer hardware and personnel cost.
Thanks for your help... Phil
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 3149.1 | who told you that? | 4GL::VANNOY | Jake VanNoy | Wed Aug 01 1990 00:00 | 12 |
Who told you the DECwindows Program Office was doing a characterization
study? Being one of the 3 people here in the DWPO, there is a pretty
good chance I would know that such a thing was happening, and no one
has told me... ;-)
There were a study done by VMS which I honestly don't remember whether
it was published or not. There was a more recent study sponsored by
the VT1000 group, and I believe I have seen that published. The VT1000
notesfile is probably a good place to look. A network utilization
study done in VMS was just published in the DEC Technical Journal issue
on DECwindows.
| |||||
| 3149.2 | Performance Reports, but still need more | DPDMAI::STEINER | Tue Aug 07 1990 14:15 | 23 | |
RE :-)
Well, it's sort of a question of viewpoint, or, which turf do you want
to defend.
The VT1000 people are not very fond of reports that claim high overhead
generated on the X-Client by X-Sessions to the X-Server with a VT1000.
The X-Station people are not very fond of the reports that claim
similar overhead for ANY X-Client session regardless of X-Server node
configuration, bur rather depending on the interactivity and image
content of the X-Session.
A Report that I've read is from Paul Antoniescu in the DECWINDOWS_5W
notes file, which claims .5VUP an 2-3MB load per DECwrite session. This
report was done by the Steve Bennett, LES Systems Engineering System
Characterization Group. For high graphics content the load could be many
times the above requirement.
Another report was on the VT1000 positioning by the Internal Product
Strategies Group.
Yet another article was by John Buford, where similar or even more
requirements were found to be necessary.
| |||||