100Mbps Home Internet Service Next Year in Finland

Listen Up writes "According to an article on CNN, broadband Internet access via cable modems in Finland will be able to hit 100 Mbps as early as 2006. That would be 50 times faster than the average broadband speeds now offered to cable TV homes in Finland. Do you think this technology has the possibility of reaching U.S. shores? Or do you think the already deeply entrenched U.S. politics are going to keep this technology from ever reaching us? There are already thousands and thousands of miles of 'dark fiber' underground around the U.S."

Comments

What is the

[mingot]

What's the upload speed?

Re:What is the

[irc.goatse.cx troll]

No, its usually the speed of the line, its just american residential has dicked us around long enough for people to expect it.

When you run a 100mbit line to your office, if you only get 90mbit upstream you complain about the SLA being broken.
When you run a 5mbit line to your house, you're lucky if you get .3mbit up.

Re:What is the

[Anonymous Coward]

Excellent perspective on the DOCSIS capabilities and limitations, vs. FTTP. There's a third distribution model that merits some discussion as well: PPP networks. I work with a carrier PPPoETTP network at work (Ethernet to the prem) which resolves a lot of the bandwidth sharing issues typically found in less sophisticated ETTP implementations (there are other issues, such as security matters especially related to directed broadcast and similar problems that PPPoE provides further control over).

Many of our customers migrate from either the big DOCSIS provider in the nearby metro (very well run, I should add), or from poorly run incumbant DSL networks lacking PPPoA. Their feedback is that the network is almost always faster than what they used to have (yet the advertised rate is often less - e.g. 500 kbps). Three things account for this, one of which is better control in the last mile via PPP protocol overlay over the DSL or Ethernet transport. Well designed transmission systems are also important - e.g. taking your traffic from the head-end of the customer connection and bringing it back to your central node. Egress capacity is the final component - e.g. where does your wholesale Internet come from.

I should note that wholesale capacity costs nothing like retail per Mbps. Anyone who compares the two is definitely dealing with apples and oranges. Wholesale IP is sold with the expectation that it will be packed - completely aggregated - by the carrier buying it. The upstream is selling you (hopefully) unaggregated capacity, meaning of 45 Mbps, 45 Mbps less protocol overhead is there for your demand. DS3 rates from reliable, unaggregated providers still exceed $10K per month for 45 Mbps flat (not a burst/measured product). That's a realistic ~$200 to $250 per Mbps per month.

So when a home user expects 5 Mbps sustained for $45, you know there's going to be a problem.

In case you're curious where the DOCSIS guys are going, expect further control and differentiation of the application protocol used. SMTP and DNS regulation were just the start (and more of a reactive action). Transparant HTTP proxy has its issues as well. Regulation of VPN protocols on residential networks was the first clue as to the long-term direction. Expect an opt-in protocol model in the next couple of years rather than opt-out (for instance, you will be given permission to use HTTP/80, HTTPS/443, and explicit "legal" ports) and be further forced into proxies of the legitimate services when possible to ensure you're not putting tunnels masquerading as legit services. The security and support issues have changed the profit model, and price competition is really heating up with WiMax coming and BPL threatening (whether or not that happens is another issue). Either consumers will have to fork out $80 or more per month, or they're going to be eating a lot less. Given the marketing pressures of "more speed is better," it's probably going to be "speed with a ton of rules."

Hope you like fast web pages and just fast web pages...

Nice!

[Sweetdelight]

I wish I lived there..And if the fibre doesn't have Caps, That would sweeten the deal.

Benchmarking

[hattan]

"Based on our research, 30 megabits per second is the absolute minimum in future homes" I wonder what kind of tests they used to determine that figure.

Meanwhile in japan ....

[mxpengin]

I have this FTH service in japan since last month and is very nice ... my only complain is that is very hard to get high transmition rates with the service... only if you are using things in japan . The cost is about 80 dollars a month and television services can be used on demand ( for a fee of course ). A link in english to my provider .

That's what I'm getting _today_.

[mdn]

Well, not really. But I could be getting it if I wasn't too cheap.

100 MBit Internet access (both ways) is offered to apartment owners in a number of Swedish towns. This costs about 76 USD a month.

As I said before, I'm too cheap to pay for that, so I'm paying for a throttled version (10 MBit/s) of the same service putting me back about 40 USD a month.

The service has been offered for quite a few years by a company called Bredbandsbolaget. (The site is in a strange foreign language though. Be warned.)

Re:100Mb/s speed

[KarmaMB84]

A couple dozen infected machines in Finland can now DOS attack multiple internet back bones simultaneously. GREAT.

They can only afford to have 300 subscribers :-)

[Anonymous Coward]

The last time I looked at the pipes going between Finland and Europe, there was only a 3Gb/s line in place. Is the Finland-only bandwidth?

Um....?

[Atario]

Aren't all connections shared, at some point? Isn't that more or less the definition of the Internet?

Those DSL lines all go somewhere, you know, and I'm betting each one doesn't get its own T3...