"HOFMEISTER,JAMES (HP-USA,ex1)" wrote:
> RE: VT-MGR vs. Telnet
>
> Performance consideration with Telnet:
>
> 1) In our example 4 times more packets.
> 2) Each packet is a CPU interrupt on the HP-e3K system.
>
> I am 95% sure this is the way it works... but I will leave a 5% margin
> of error to give one of you diligent folks the incentive to take a TCP
> trace and do a packet by packet compare of the 2 protocols.
Close enough for government work <grin> but one other exception - If the
client side supports Nagle-style transmission deferral (see RFC896).
Unless you are a real slow hunt-and-peck typist, you will usually have
many of the input packets collapsed into fewer, longer packets. An
excerpt from RFC896:
> The second case to examine is the same Telnet test but over a
> long-haul link with a 5-second round trip time. Without any
> mechanism to prevent small-packet congestion, 25 new packets
> would be sent in 5 seconds.* Overhead here is 4000%. With the
> classic timer scheme, and the same limit of 2 packets per second,
> there would still be 10 packets outstanding and contributing to
> congestion. Round-trip time will not be improved by sending many
> packets, of course; in general it will be worse since the packets
> will contend for line time. Overhead now drops to 1500%. With
> our scheme, however, the first character from the user would find
> an idle TCP connection and would be sent immediately. The next
> 24 characters, arriving from the user at 200ms intervals, would
> be held pending a message from the distant host. When an ACK
> arrived for the first packet at the end of 5 seconds, a single
> packet with the 24 queued characters would be sent. Our scheme
> thus results in an overhead reduction to 320% with no penalty in
> response time.
So, if you can type faster than your round trip time, it will reduce
the packet count by default if Nagle-type queueing is supported.
This applies only to standard telnet connections, however. Connections
using "local-echo" (QCTerm's "advanced" telnet) already avoid the single
character packets.
Jeff
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
|