[comp.dcom.lans] Ethernet bridges and T1-izers

kwe@bu-cs.BU.EDU (kwe@bu-it.bu.edu (Kent W. England)) (07/26/88)

In article <Jul.22.20.57.24.1988.24248@athos.rutgers.edu>
hedrick@athos.rutgers.edu (Charles Hedrick) writes:

>To use a whole T1 circuit with a bridge or router, you
>have to use a box that takes a generic 1.5 Mbps signal and puts it
>into the funny T1 format.  I call such a box a T1-izer.  
[...]
>In general the bridges and routers don't know and
>don't care exactly how fast a line is.  They just lock onto the clock
>that comes from the T1-izer of multiplexer.  

	That raises some interesting questions.  Depending how "smart"
the T1-izer is in doing the required bit stuffing, it might present a
higher or lower clock rate to the bridge.  What differences are there
among T1-izers and do vendors advertise what the true clock rate to
the interface is?

	Are any T1-izers capable of data compression and if so what is
the expected gain?  Could we get effective throughputs of 2Mbps over a
T1 circuit? Or is this capability best left outboard of a T1-izer?

	Kent England, Boston University

hedrick@athos.rutgers.edu (Charles Hedrick) (07/27/88)

Many T1-izers have options to select how much encoding is needed.
Generally the specs tell you exactly what clock rate they present for
each combination of options.  E.g. there are options to say whether
your telco well let you get away with bipolar violations (one specific
kind of bipolar violation can be used to provides ones density more
efficiently), and whether your signal has any known properties that
will allow more bit-efficent processing (e.g. ours will do a better
job if the signal is known to be HDLC).  I think in some you can
specify whether it has to worry about ones density, whether the signal
already has T1 framing, etc.  Typically you can get from something
like 1.528Mbps to 1.544Mbps depending upon the kind of coding the box
can do and what options you set.  I haven't seen any that do
compression.  We're talking about slightly jazzed-up CSU's here (the
official product name is normally "clear-channel CSU").  I think
compression (particularly at this data rate) takes a bit more
processing power than they have.