>X.25 is design for speeds of 56K or below. In a previous life (Apple
>Computer), I had lots of experience working with their X.25 network. 56K
>was/is the top end. Sorry!
>
>Darren
Not true. While the retransmission defaults for X.25 tend to be
optimized for around 64 Kbps and an analog-style error rate of around
1 in 10**5, there is no reason it is limited to 56 Kbps. At higher
speeds, it may be appropriate to tune the packet and window sizes.
There remain niche applications in which X.25, either the full three
levels or just LAP-B, remain useful. Long-delay satellite links tend
to like LAP-B. I've also found it useful to use the full three
levels on very bad links in developing countries, in order to force a
128-byte MTU that was optimal for link-level retransmission on very
noisy lines.
On the other hand, I've personally run X.25 with individual VC's at T1 rate.
X.25 throughput optimization can be complex. It involves not only
the window and packet size, but often is influenced by the number of
VC's per access link.
_________________________________
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
