Re: (ASCEND) Re: Clocking question

[Peter Lalor <plalor@infoasis.com>]

>> You need to take timing off a PRI, then have the telco have the rest of the
>> circuits take timing off you ("loop-timing"). You may be able to get away
>> with not doing this if all your circuits are from a single telco, but when
>> you start terminating circuits from multiple vendors in a single chassis it
>> becomes critical.
>
>I don't think most telcos are willing to do this :-(

"Willing" and "telco" in the same sentence? Stop it, you're frightening me.

In any case, we've managed to get Pac Bell, MCI Worldcom and ICG to do it.
We terminate circuits from all three into a TNT, and sorting out the timing
was the solution to several weird problems.

>As I understand things (I may be wrong), they have good reason not to.
>If your PRI terminates in a Mux of some kind (say, to backhaul it OC3
>or OC12 into the CO), the Mux probably does not have pleisiochronous
>hardware (i.e. it cannot buffer and adjust the signals to match the
>various clock rates coming from its various input and output ports), so
>it is probably forced to clock off its aggregate port (to the CO) and
>supply the clock to all the other ports. And the aggregate port clocks
>off the CO's redundant master clock at the other end. If your PRI goes
>straight into the switch, I still don't think you could necesarily do
>it, or, at least, you'd have to have a common clock on all the ports
>on a single linecard (how many ports are there on a DMS 100 T1 linecard,
>anyway?).

FWIW, we colo that TNT with MCI Worldcom, so they cross-connect everything
from everyone. Don't know if this affects clocking or not. I think not.

>Am I way off?

I sure hope you're not asking me. You used "pleisiochronous" in a sentence.
if this were Scrabble, I'd forfeit.

Peter Lalor           Infoasis
plalor@infoasis.com   http://www.infoasis.com/
++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>