Canada joins the 21st century !

Blake Hudson blake at ispn.net
Thu Dec 22 16:32:21 UTC 2016


Jean-Francois Mezei wrote on 12/22/2016 8:59 AM:
> ...
>
> Yesterday, the CRTC declared the Internet to be a basic service (which
> enables additional regulatory powers) and set speed goals to 50/10.
>
> Note that this is not a definition of broadband as the FCC had done, it
> one of many criteria that will be weighted when proposal to get funding
> is received. But hopefully, it means the end of deployment of DSL.
>
> ...

Some rural areas in the US are seeing either VDSL2 or bonded DSL 
deployments which do push the capabilities available via copper to the 
home deployments. Having operated cable, DSL, and fiber networks, I can 
say that cable and DSL have managed to stay relevant a lot longer than I 
expected.

I also think it's a bit disingenuous to say that cable provider meets 
the 50/10 standard and DSL provider doesn't when a hypothetical cable 
plant might provide 1Gbps x 72Mbps of last mile bandwidth shared between 
100+ subscribers and DSL plant might provide 20Mbps x 1.5Mbps dedicated 
to each subscriber circuit. In this example, the worst case for the 
cable plant is 10Mx768K per subscriber (note, this doesn't meet the 
specified goal) & the DSL plant is 20Mx1.5M per subscriber (twice the 
speed, but also doesn't meet the specified goal). In other words, 
sometimes the cable subscribers might experience faster speeds than DSL 
subscribers, sometimes they might experience slower speeds. The DSL 
subscribers would see consistent speeds.

If a provider's deployment model provides "up to" the specified 50/10M, 
but subscribers see lower than these rates, then I would argue that the 
provider hasn't met the goal. While a cable provider could engineer 
their plant to dedicate 50/10 bandwidth per subscriber, this is not done 
today and will not be done in the near future because it isn't 
profitable (whether its necessary is another discussion). This seems to 
be profitable in FTTH and vDSL2 deployments because providers are 
exceeding these goals today in the US.

Both cable and DSL technologies have managed to stay relevant longer 
than I expected because each seem to have pros, including the ability to 
utilize an existing infrastructure. I don't see either technology dying 
immediately, and can imagine that vendors may yet be able to eek out 
more performance from the existing infrastructure, but I do see more 
future potential using fiber to the premises infrastructure.


More information about the NANOG mailing list