Subject: | |
From: | |
Reply To: | |
Date: | Tue, 3 Sep 1996 15:02:07 -0500 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Lars Appel said in commenting on WRQ and MPE/iX 5.5:
>Even with inbound telnet it might still be a better choice to use NS-VT as
>the protocol for HP 30000 sessions as it is more efficient for the system
>as well as the network, as far as I know. I think, telnet sends a network
>packet per character (by default) whereas NS-VT sends a packet per line.
WRQ released a white paper at HP World on the comparison between Telnet and
NS/VT throughput. We've got a report about the paper in the September
issue of the NewsWire (now online for paid subscribers like Lars at
http://www.3k.com/other/seybold/subscribers/Alonline.html). In short, you
can expect a lot more overhead on simple commands (40X as much), and
somewhat more overhead on things like LISTF.
WRQ's tests were with the Telnet Access Card. Telnet/iX Server is expected
to be slower, but nobody knows by how much just yet.
Telnet access, while universally accepted across platforms, is another
example of the hidden costs of "open systems." NS/VT, like IMAGE, is a high
performance tool that will always outperform such industry standard
solutions. Balance your performance against connectivity and productivity,
as always.
Ron Seybold, Editor In Chief
The 3000 NewsWire
Independent Information to Maximize Your HP3000
[log in to unmask]
512-331-0075
|
|
|