Slashdot Mirror
South Korea Plans National 100 Mbps Network
prostoalex writes "Korean Ministry of Information and Communication is planning to wire the entire country with high-speed 50-100 Mbps network. A total of $80.4 billion will be spent on the project that's expected to be completed in 2010."
You mean when "high speed" isn't high anymore?
Don't quote me on this.
If you think about it, that's about $2000 per Korean, perhaps $6000 per household for high speed internet access, which won't be complete for 7 years. Are they really getting a good deal?
This is an interesting approach to infrastructure. Now, the next question: how will this approach affect Korea's economic development? What types of businesses will get located in Korea specifically to because of the ubiquitous availability of this type of infrastructure? How will the universal availability of broadband affect Korea's land use of development patterns? Will folks still commute via cars? Will factories start to become remote controlled?
Throughout their country, they have true 10mbit connectivity for most of its citizens at roughly 13$ US per month. That is insanely awesome when you figure in the fact that here in the US, we pay around 45$ to get at best 2mbit connectivity that peaks out at right around 140k most of the time. And thats just downstream.
While technology is increasing rapidly enough to make local network connectivity at extremely high speeds economically feasible for the first time, WAN technologies are still another story and lag behind by a few years. You still want dedicated 1.5mbit connectivity, you are STILL looking at around 800$+ dollars a month. (Key word being dedicated).
Good for the S. Korea!
hrrm.
If you look at the numbers their complete budget for 2000 was only $95.7 billion. Assuming it starts now & ends on time, without any cost overruns, we're looking at something like 12% of the government's spending going towards this project.
That's some commitment to closing the 'digital divide'. Well, as long as they make reasonably affordable computers available to their citizens when this thing goes live.
my sig's at the bottom of the page.
Okay, for those of you already piping in that this isn't as fast as it can get, I'd like to see your present hi-speed home access push far over 1 Mbit/sec. Nevertheless, this begs the question:
In 2010 will 100 Mbits be considered fast or slow? Is there a "Moore's Law" for Internet access speeds? Back in about 1982 I was connecting to the local BBS with a 300 baud modem. A megabit download speed (today in 2003) is roughly 3000x that speed, and we're there after 20 years. That equates to almost exactly a 50% increase in speed per year. So if we go another 7 years at that rate, by 2010 we would consider 16 Mbit/sec to be fast.
Okay. I'm envious.
Murray Todd Williams
NTT and other companies have already been offering 100Mbs fiberoptic lines to homes in Japan for quite awhile now.
The best part is it's cheap,
They usually cost a little more than $40 a month.
Of course, it's still twice the price of 12Mbs ADSL lines in Japan like Yahoo BB who offers 12Mbs speed for $21/month. Most people don't know what to do with 100Mbs anyways.
...and not to be out done, North Korean President Kim Jong Il has asserted that his country can compete with the decadent capitalist South by establishing the Socialist Communication Organization (SCO) to provide tin cans and string to 1 out of every 100 loyal members to the party.
-Crolis
I can't believe they couldn't find a better use for all that money. High speed internet shouldn't be something that is critical in a nation that still needs much development in basic infastructure. For that much money in the US we could do so much it is beyond most people's comprehension.
The only justification I see this having is the 370,000 new jobs, but how temporary are those jobs. Will most of them disapear after the system is put up and there is nothing left to build let alone money to build it with. To learn more about what we in the US could do with $80 billion(around what is being spent in Iraq go here
If we need it for such basic things I would think a less developed county would need it even more.
Why not just wire them onwards to consumers' homes?
Because Fast Ethernet switches are chump change, and fiber switches cost more than many people's houses.
Optical switches are designed for backbones, not connecting everyone and their dog. DWDM, Sonet and ATM don't easily (or affordably) scale out to many-2-many connections.
Learning HOW to think is more important than learning WHAT to think.
Cost and switch fabric.
All these posts who talk about 1 Gbps and fiber aren't thinking it through. The difficulties and costs aren't associated with the cabling or end-point connections -- they're at the switch.
1 Gbps is nice. Now pump an entire apartment unit with GE into the switch. What speed will the internal switch fabric have to support? Assume 200 apartment units, then that is in the neighborhood of 200 Gbps of switch fabric throughput. Consider most of the traffic will be going OUT of the building, the outside pipe will have to be something like an OC-48 ATM or 10-G ethernet connection.
Now THAT switch, and 1,000 more like it, all feed into different switches and the problem multiplies.
Think of the RAM buffers, latency and clock frequency that has to be maintained in the switch to handle 200 Gbps of thruput.
Cisco's top of the line Catalyst 6500 series boasts:
# 32-Gbps bus--Allowing access to a central shared bus
# 256-Gbps switch fabric--Located on the switch fabric module (SFM)
# 720 Gbps switch fabric--Located on Cisco Catalyst 6500 Series Supervisor Engine 720
So you ARE pushing the edge with mass deployment of fast ethernet.
Oh, yeah. Fully loaded 6513s run $100,000, easy.
Learning HOW to think is more important than learning WHAT to think.
That's a great misconception. Quantity does not always equal quality. They are competitive, but overall, no better than Europeans or gamers from other Asian countries. WCG has them at #3 in the final medal tally.
As for the topic at hand, good for S.Koreans. It's nice to see a nation thinking forward and wiring it's population with a forward thinking attitude. The government has its finger on technology's pulse. Unlike in US, where we still have anti-competitive carriers/ISPs monopolizing regions and not getting reprimanded for it. We, as Americans, are behind the progress curve, in terms of broadband connectivity. There is a huge chunk of population still using 56k modems. I mean, Christ. Broadband should be a cheap commodity and a requirement in every house.
While Asia, Europe and other continents are focusing on the future and doing something about it, our politicians are dicking around with special interests and not thinking of the implications 10 years down the line.
South Korea: Lets build a grid and give every citizen access to broadband.
United States: Let companies decide instead of the consumers. Profit comes first.
South Korea: Technology is the future. Internet access is a basic human right.
United States: Intellectual Property is being violated, lets greenlight tyrants like **AA to set the agenda. MP3 Downloading has to stop.
South Korea: Open Source in Government? Lets keep our options open.
United States: Let Microsoft get away with everything, as long as they contribute to political campaigns.
As you can see, we'll be still arguing about having pioneered the Internet and other technologies in irrelevancy, while other nations surpass ours and make the rules.
I'm amazed at the number of poor posts that get moderated up whenever there is a telecom related article.
/. Reader has a ADSL line and got lucky with no neighbors using outflow bandwidth and an ISP that doesn't care (yet), so therefor any nonsensesical pronouncement that would lead to the whole world having a service that now costs $5,000/month being provided to them for $21.95 makes perfect sense.
I just scanned through the two dozen that made a +3 or better so far and I'm astonished at the number of poor assumptions about physics, economics, network operations, and life in general.
The physics was the most egregious of the bunch and I think everyone who is smart enough to navigate far enough to see this *should* understand, but I can't resist brushing some of the others.
Moore's law is just an observation - its *NOT* a law. Why is someone applying this to available circuit speeds for WAN access? WAN access lines are very expensive and thusly that ground has been throughly worked by every telco equipment vendor - copper pairs are good for a about 2 mbits at the typical distance between a home/office and a CO, the next step up is DS3 delivered on coax (low loss, damned expensive compared to copper, and fiber refits in existing areas are crazy expensive. If it was possible high value DS3s filled with 672 voice channels would be the first thing going on some new wonder technology - this isn't happening, ergo it doesn't exist.
And why are they making statements like "100 mbit stuff is cheap on ebay, just build a national network out of it". Ethernet is a *LAN* protocol - 300' limit in most cases for copper, Cisco 2950-LRE are only good for a few thousand(hint, long reach ethernet == DSL), and who would want to manage a pile of crap from ebay? The number one expense in any network operation is almost certainly payroll and a crapola network guarantees 127% of revenue will be spent unfornicating it. If you want reliable service you pay for reliable gear. Once in a while you get lucky on the cheap but no business big enough to do a neighborhood size rollout would fool around like that, let alone a big telecom organization.
It seems to me the underpinnings of many of those posts are pure emotion coupled with a sense of entitlement - J Random
Mod me brilliant, mod me troll - the opinions of the readers are foolish and the moderators deserve a timeout for promoting such crap.
I am very easy to get along with, but I don't have time to waste being nice to people who are being stupid. -Theo
No, it won't be obsolete quickly by any means. The reason is similar to why Pentium systems above 200Mhz don't go obsolete, they're sufficient for audio and video which makes them entertainment devices rather than strictly computing devices. People often keep televisions and radios for decades.
100Mbps is fast enough to stream not just full bitrate Mp3s, but decent quality video as well. So, it might not be the fastest forever, but it won't be obsolete for a long time.
I'm not going to get into whether or not the country should spend that much money on the network when it has many other problems, but...
People saying 100Mbps won't be fast in 7 years? Screw that. If you think we'll have even 1/10th of that in even 1% of the US in 2010 you're out of your mind. Huge areas of the nation don't even have 56k-capable telephone lines, let alone broadband. This won't change until it's profitable for the businesses to do otherwise. Monopolies own all the lines, and there is no government incentive. There won't be, either. (Which is good and bad)
I've got 1.5Mbps right now, with planned 3Mbps in a year or so. I've only had it for a few months. I don't see it going up much more by then, considering how long it took me to get above dialup...and certainly not to or above 100Mbps. Hell I bet 20Mbps will be a lot in 7 years if you live in the states and we're talking average residential internet speeds. Same goes for globally.