Hang on folks.
The bottleneck limiting bandwidth growth is actually routing/switching
capacity, not bearer capacity.
Even if you do succeed in getting the bandwidth between you and the
exchange, using whatever technology, the problem is what to do with it at
If you have a couple of hundred 45Mbps links (or even a couple of thousand
2Mbps links) running at anywhere near capacity, you will have 10's or 100's
of gigabits of traffic to deal with at the head-end. You just can't route
that, or anything like it, with today's technology. You'd have difficulty
even switching it. That's the equivalent of 100's of thousands of phone calls.
This is the experience with ADSL in the 'States. You get fast transport to
the exchange, but if you only have domestic-rate ADSL, you get pretty
crappy throughput from there onwards.
Because of anomalies in the money-sense continuum, there are a lot of
Internet businesses where the conventional rules of business do not apply.
Sadly, however, the rules of physics still hold through even around the
most pronounced of these.
At 03:32 PM 3/1/00 +0000, Justin Mason wrote:
>>Mike Knell said:
>>> Hum - the concept of "the net" as a creature of infinite bandwidth that
>> is limited only by your provider's connection to it is one thing, but
>> in reality the Internet is a congested mess of fibre and more-or-less
>> working backbones that can't expand fast enough to cope with the
>> clouds of HTTP segments that fill it up. If everybody suddenly got
>> 45Mbps of bandwidth to their house, this doesn't mean everyone would
>> be able to download anything from anywhere at 45Mbps. The core network
>> just couldn't cope with it. Sure, you'd be fine for local stuff, and
>> things would be generally faster, but all your internet use would not
>> suddenly become 720 times faster.
>> A network is only as fast as the slowest point in it, which in this case
>> would be, well, most of the network beyond your local one. There's more
>> bandwidth in the big backbones, yes, but it's shared between a lot more
>> people and a lot of major networks. It'll improve your latencies, sure,
>> but overall transfer rates won't get 720 times faster. They'll get faster,
>> but not that much faster..
>>Yep -- there's been several articles in US mags along these lines in the
>last few years; people reviewing new whizzy high-speed access, and
>basically saying "well it's slightly faster than it was before". (I think
>Boardwatch would be a good spot to find some of these).
>>Personally I think it will be more revolutionary to have always-on
>internet at home, than high-bandwidth internet at home. Next benefit
>would be latency reduction. Finally, an increase in available bandwidth
>is a benefit -- but not the be-all and end-all of it.
>>Also, I think that wide availability of high-speed home access through a
>shared ISP's backbone will allow more peer-to-peer stuff; it'll change the
>normal use of the internet from the current "dialup users reading
>webpages" to something new... could be games, I'm sure there'll be other
>new uses. If it is games, then the latency reduction will certainly
>revolutionise that area. One thing's for sure -- people's use of the 'net
>is a moving target... witness the sudden explosion of HTTP after Mosaic
>>> > Someone on ILUG mentioned that the Genesis thing sounds like VDSL (see
>> > http://www.dsl.com/intro.html#22), which is very high bandwidth but
>> > very short max cable length (3000 ft apparently). So it does exist --
>> > maybe only in a lab somewhere -- but it is out there!
>>>> Getting stuff working in a lab is all well and good, but getting it
>> working in the real world is something else. I'll remain happily
>> sceptical until it's published stuff rather than "sorry, can't talk
>> about it" right now. Secret technologies are all too often snake oil.
>>I agree with you on this one -- snake oil, or just pure
>>> underlying infrastructure has speeded up a lot. Many users (not anybody
>> in this thread, I should emphasise, it's merely a Slashdot-reading
>> observation) seem to assume that high-bandwidth Internet access with
>> no limitations is a right - see those US students threatening to take
>> universities to court because they're blocking use of Napster on their
>> _academic_ networks. Personally, I just think that's greedy and shows
>> some pretty fundamental misunderstandings of the realities of today's
>>high-bandwidth Internet access may not be a right -- but it's something to
>wish for... ;)
>>>Antoin O Lachtnain Phone +353-1-676-8996
Nua Ltd. Fax +353-1-2839988
Merrion House email antoin at nua.ie
Ireland Nua: New Thinking for the Digital Age.
Maintained by the ILUG website team. The aim of Linux.ie is to
support and help commercial and private users of Linux in Ireland. You can
display ILUG news in your own webpages, read backend
information to find out how. Networking services kindly provided by HEAnet, server kindly donated by
Dell. Linux is a trademark of Linus Torvalds,
used with permission. No penguins were harmed in the production or maintenance
of this highly praised website. Looking for the
Indian Linux Users' Group? Try here. If you've read all this and aren't a lawyer: you should be!