Slashdot Mirror
Why We Don't Need Gigabit Networks (Yet)
AmyVernon writes "Most computers today can't support gigabit connections and current Wi-Fi networks can't offer those speeds either. The first trial of Sonic.Net's gigabit network was a speed test on a generic laptop that showed off 420 Mbps down; the laptop couldn't handle a full gig. Plus, few applications need those speeds. It's hard to justify such a huge investment in a network that will have few subscribers and few applications that need it. Of course, that can change, and then these networks will be vital. This story has a good analysis of where things stand and what has to change."
Gigabit networks are important when working with almost any kind of file copy. I am not sure the last time someone tried to backup even just 100GB of data (Think backups) over a 100 megabit network. Copies like that can take for ever a fully saturate 100 megabit network and slow down traffic for everyone. While copies over gigabit rarely use the entire pipe its good to know that there is still bandwidth left over for other tasks.
How DARE you say we don't need faster networks! This article should be purged from the interwebs and timothy should be strung up by his gonads for even considering posting it!
I am not sure how "one does not need a gbit connection". Even a small file server in 2006 could output more than 70MB/sec (practical test on large files).
If we started today, by the time we build out a gigabit-to-the-home network, PC components will easily handle the speed. As to whether we need it, we may not need it now - but we'd quickly find a use for it if it were available.
420 megabits per second is still over 4 times faster than 100 megabits per second.
Stop thinking "need" and start thinking "it's cool". I don't "need" a new computer right now, I
would do fine with this laptop at least 2 more years (already 2 years old). Still I'm buying one.
Where would we be today If everyone was just thinking about what they actually needed?
I actually routinely do file transfers to RAID arrays where the bottle neck is my gigabit network. While I agree with you that the general population doesn't need this level of speed, what about even a small business network with a heavy network load?
I have 2 computers, a ps3, and a wii connected to the net. even if I am doing something simple like streaming a movie from one computer to the ps3 to watch on my tv while someone else is playing a game online, downloading something or the other, or just generally using the web to watch anything in HD, I could easily find a use for that bandwidth.
I think the gist of the article is that we don't need gigabit connections from an isp to the home. While that's debatable, that's not the same as saying that gigbit networks within a home or office are not necessary.
Well.. maybe. Or Maybe not. But Definitely not sort of.
"14 Mbps ought to be good enough for anybody."
What do you mean "speed test", we've had gigabit broadband for some time in Sweden. It costs about 900 SEK/€100/$140, you can order it here: https://order.bredband2.com/index/products/cOrderType/broadband/iCitynetID/768537
Not available in all cities of course, but still. Not much news with gigabit broadband.
And that WiFi and most peoples computers, let alone routers are unable to push those speeds is not newsworthy either. At least not at /. ...
Sure, connecting to a text based BBS didn't need broadband, but looking back, I'm pretty happy we all upgraded from our 300 bps modems!
The reason this post is stupid is that infrastructure is long term. When you go to the trouble of sending out a crew to dig up and put fiber in the ground your putting in an infrastructure asset that should have a 15-30 year lifespan. The fact that can average machine can't saturate it today means we're being forward thinking.
... and 640K computers.
Build a faster network and someone will invent more devices to connect to the network to shove around data that they don't need.
Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
The problem is that TFA was not correctly edited.
1. She's talking about gig connections from your home to your ISP.
2. She's mixing wired and wireless.
3. She mixes gig and 100Mb/s.
4. $40 for 100Mb / $70 for gig is NOT a lot of money.
5. She's wrong. Computers today CAN handle a gig connection.
6. So what if the cheap router/firewall/whatever you have cannot handle a gig connection (it can probably handle a 100Mb/s connection)? That's the easiest piece for the consumer to replace.
7. The apps that would use it TODAY are things like streaming media. Getting the WHOLE movie or song or whatever 100x faster means fewer delays from the consumer's point of view (perception).
If I could get gig speeds here in Seattle for under $100 I would certainly do it.
not everyone keeps/needs a server at home.
The misconception that people would have no use for a home server has led to prevalence of highly asymmetric Internet connections suited more to a spectator culture than to a participatory culture. Have you heard of the FreedomBox?
TFA is actually about Gigabit connections to the internet. Not local infrastructure.
I could understand the attitude if it would be applied to ten or 40 gigabit Ethernet. When talking of gigabit networking, I see this to be absurdly badly timed. Broadband speeds above 200 Mbps are widespread in developed countries (I'm not certain if US qualifies), and infrastructure upgrades make true gigabit-to-home more and more viable. There isn't much point opposing using gigabit Ethernet at consumer side (GE ports cost peanuts these days), and if consumer is ready to pay, there's no reason not to handle the capacity also on the last-mile link.
If one throws idea of subsidising such last-mile technology, it's a bit different question, though, but why oppose the technology in general?
Plus, few applications need those speeds.
That, I can agree with. How many high def uncompressed live video feeds can a household watch?
For example, ATSC "over the air HDTV" is only 20 megabits/sec, so I could watch 50 HDTV channels simultaneously...
"Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
My house has been wired with cat5e since 2002 and it's been connected via gigE switches and NICs since that year or 2003. Even last year's netbooks have 1000BaseT support. Desktop motherboards have supported gigE since NForce3 was new. Only my cheapest socket 775 systems don't have gigE on the mainboard and that can be remedied with a PCI or PCI-e NIC.
FTA:
"So we’re stuck at a point where a gigabit — or even 100 Mbps – sounds awesome, but it’s not exactly worth the prices most companies want (or need to charge). This is why Google’s and Sonic.Net’s plans to expand moderately priced 100 Mbps and gigabit networks will be so important."
The summary to this article is misleading. It led me to write a mini-rant about the usefulness of gigabit LANs. In fact, the article's talking about gigabit WAN connections at the home. Their denouncement has the tinge of that old Microsoft exec quote about the internet being a fad and no one needing very much ram.
FTA:
“If every consumer has 100 Mbps, we’d have some better applications,” Jasper said. ” At 100 Mbps, high-def video conferencing becomes a reality and you don’t need local storage anymore. You don’t even need local computing.”
You went from talking about gigabit WANs (at the corporate level), to the use of fast ethernet WAN at home. Somehow, there's a use-case at the home that isn't there at the corporation.
And this made /. frontpage, why? Can I get a +5 comment simply by using the words "100 mbps, gigabit, ethernet, 802.11[n-z], important, high-def, local, storage, computing" ?
PS: I don't reply to ACs.
is not for one computer to saturate it, but for 10 machines to get decent throughput simultaneously.
As the island of our knowledge grows, so does the shore of our ignorance.
"640k ought to be enough for anyone." "There is a world-wide market for about fifteen computers."
Time and time again people have been deeply mistaken about anything having to do with the future of computing. The first time I saw a VGA display I was so smitten that I thought "this is the best it can be". Well I was wrong and so were a lot of other people who thought that there would never be a need for something more advanced than what technology has to offer today.
We expand technology by pushing against the current limits, finding things that cannot be done as the technology exists today and figuring out how to get it done and then proposing that solution as a standard. Meanwhile someone else has found a different solution and they too have submitted it as the standard. Then the wrestling begins perhaps one solution is accepted or maybe the solutions are married or an entirely different standard is found or maybe everything is rejected and they become proprietary products.
I'm 55 years old with lots of gray hair yet I still hold out hope that before I die I will be able to play, just once, in a holodeck, maybe with someone overseas. That could require something more than gig lines.
Shared networks work best if they are unsaturated. Having a very fast network is a simple way to achieve that without having to sweat bullets over how to enforce fairness in the sharing algorithms.
Getting the WHOLE movie or song or whatever 100x faster means fewer delays from the consumer's point of view (perception).
From the consumer's point of view, what's the point of downloading faster than one can listen? I can see only two reasons: 1. to skip around in the file, which could be handled with out-of-order downloading techniques such as HTTP range requests; or 2. when a handheld media playert will soon be moved out of reach of a fixed Internet connection (the download at home and watch in back seat/bus/train/plane scenario).
The title says "Why We Don't Need Gigabit Networks" ... whilst the article talks about Gigabit Internet...
About "networks":
A laptop with its integrated network chip might have trouble processing 420Mbps. A Gigabit network might have trouble spewing more than 500Mbps in 1500-bytes MTU TCP there the round trip takes a relatively hefty sum of time.
But 420Mbps versus an approximate 80Mbps at FE can spew is still more than 5 times faster. And whenever you need to transfer files, it does count abominably. It means I can do my 5-50 gigabyte backup in minutes versus hours. It means I can safely transfer some HD movie to my PS3/360/Mac without a hitch, with proper buffering, instead of being stuck at the limit of my network. And imagine if my automated backup decides to start at the same time I'm playing a movie... in FE instance, that would mean movie skips.
So for networks, it makes total sense.
About "internet": ... Not only are we able to transfer files quickly, not losing time, but the whole deal about being stuck waiting for a transfer to be done means waiting for minutes instead of hours. It also means users are free to roam the Internet without limitations, only "normal" things (no infringements, nothing illegal, no X, 500MB+ file transfer => tell us or you might get disconnected).
I've had the luxury of setting up a real full GE network for my job. And boy is it worth it! (FWIW: we paid for full duplex GE burstable, and full duplex FE average)
It also means we have the luxury to set up whatever server we want in-house, and not having to pay for additional servers through other means.
So for big companies, it's great. For small companies, it's great. For one individual user, it's not really useful, but it could be made very useful by having a small CDN farm inside the provider, with boxes from the major CDN players, and tell any transfer originating from that part would be sent at unlimited speeds without bandwidth usage. Not only would it save providers a lot of bandwidth, it would save us from transferring gigabytes of software updates at subpar speeds. It wouldn't be deep-packet inspection, speed bumps, transfer size slowdowns or other provider nastinesses, but an actual added service, on top of your default offering. And having these transfers done at full GE speed would mean the WORLD of difference.
So, just because a single "generic laptop" was sold with a crap GigE card then no one should be able to get such a connection?
It looks like a desperate attempt to bullshit their clients into believing that they want what they really want but the ISPs don't want to provide, and instead what they really want is what the ISPs is already providing. It's the "640kb is enough for everyone" shtick, but bullshittier.
Slashdot, fix your code or at least hire someone who is competent at it to do it for you.
When you wake up to the obvious facts of 1999, let me know, and I'll give you an invite to the 21st century. Cuz I'm k3vvL and rollz like dat.
Shoes for Industry. Shoes for the Dead.
So useing 1 system and a laptop for that is a poor test.
Why not test a desktop system? A system with a SDD?
Test with 2 systems on the same link at the same time?
There's no point in having gigabit networks, network cards can't handle that traffic.
There's no point in having network cards handle that much data, networks don't have that kind of capacity.
Why the hell do you think you're entitled to say what we need and what we don't need?
Idle, anyone?
The streaming model can die the death it deserves at that point.
Streaming is still the only model I can see for live events such as news talk shows, sports, scripted sports (e.g. WWE PPV), concerts, and the like where viewing begins before the whole video has even been recorded.
Time to recycle those old Novell Netware licenses !
The Cloud - because you don't care if your apps and data are up in the air.
"I think there is a world market for maybe five computers." - Thomas Watson, chairman of IBM, 1943
Most of the systems out there already download faster than you can watch / listen to the content.
But they still have issues where there are delays and the play-back has to pause and "buffer" more content.
Simply put, the longer the download process is (all the way down to receiving the packets a microsecond before playing them) the more likely it is that something will cause packets to be lost or delayed and the system will have to interrupt the customer and "buffer" more data.
With higher speeds, if you can download a 2 hour movie in 10 minutes you won't see any Internet-based delays for an hour and fifty minutes. Which makes for a better "customer experience".
100 Mbps and 1,000 Mbps costs the same. Both require FTTH and the expensive part is the fiber. The equipment to run gigabit on that fiber is almost the same cost as 100 Mbps equipment.
Gigabit internet is also not expensive. It turns out that most people do not use huge amounts of bandwidth just because it is possible. They will take advantage of faster download and upload speeds. They will do offsite backups. But since that backup now is 10 times faster it takes 10 times shorter. When you are done, someone else can use the bandwidth. With enough people it evens out.
I live at an apartment complex with a shared gigabit fiber. We have about 1000 active subscribers on that fiber. The average bandwidth used is approximately 100 Mbps and peak is about 300 Mbps (measured in 5 minute intervals). We never get near the full gigabit even though it is available. This does not mean it is useless. It is very useful to be able to download that ISO and not have to wait.
There are lots of use cases and more will come by each day. The major blocker right now is not that my computer is too slow, but that the servers out there usually are not able to deliver at those speeds. That will change when more people are able to take advantage of greater speeds.
Our apartment complex was able to buy a bulk gigabit fiber and share the cost. If you want to buy gigabit, even if FTTH is available in your area, they will rape you hundreds of dollars a month. Do not let this fool into thinking this is the real cost of gigabit internet. This is just what they can get away with because there is almost no competition on high speed internet. There is an issue if someone tries to run a commercial high bandwidth service on your gigabit offering. But for most customers offering gigabit is not one cent more expensive than offering slower speeds.
There are about 200 people who use my department's network at any given time during the day, and maybe 50 at night. All the desktops have their filesystems mounted on NFS, and people routinely upload or download large datasets. Gigabit networking is not even fast enough for what we do (yet somehow we have trouble getting that much installed).
Palm trees and 8
And other bloated tech.. and we will never need it.
---- Booth was a patriot ----
GigE comes on nearly every new computer. Consumer GigE routers and switches are marginally more expensive than their 10/100 equivalents. Most home media appliances are GigE. Nearly any home NAS you can buy has GigE. Most Cat5e cable can handle GigE speeds. Why would you bother buying 10/100 equipment? Even if your ISP isn't yet capable of it, there is plenty of application for it in the home, and it's conceivable that consumer-grade services provided by US ISPs will break the 100Mb barrier in the next five years, which isn't an unreasonable life expectancy of any of the aforementioned equipment. Say nothing of the benefits of GigE in small business and enterprise.
Flatulent, unwashed, blind and deaf elephant in the room is more like it. Not to mention a dumb article.
420 Mbps is more than 100Mbps, so while that laptop may not be able to use the full gigabit, it can saturated a 100Mbps, which validates the use of gigabit network.
If God had intended man to fly, he would have given him wings.
Seven puppies were harmed during the making of this post.
From where I sit on a consumer cable connection it isn't about max speed, it's about being able to carry the current customer load and no caps.
So if a Gigabit network would solve or at least alleviate those issues then it is justified, the corps can't seem to figure out a way to charge you more for giving you what they initially promised to start with, so apparently it isn't justified from their point of view.
"If any question why we died, Tell them because our fathers lied."
This combined with how long it will take to layout such a network. It won't be completely rolled out nation wide by time computers begin catching up on the higher ends anyways. Any new network laid out today would be obsolete by completion if built to today's high end or average capabilities.
With faster networks you can download an entire HD movie is much less times than it takes to watch it. By clearing the intertubes of so many low speed connections it makes it easier to route information through it. Faster networks makes so many things that are difficult now trivial. Try providing decent network access to a small apartment building or teleworking with largish datasets. 640 kbps is all anyone should ever need.
As a Virgin Media customer, I need gigabit internets as the upstream will be ( if the ratio is similar to my current service ) 60Mbit.
They whose government reduces their essential liberties for temporary security, receive neither liberty nor security.
...you don't need something is usually looking to take something away or prevent you from acquiring it.
When Fascism comes to America, it will call itself Anti-Fascism, and tell you to give up your guns.
I remember reading a PC Magazine review back in early 87 or so, talking about the recently released Intel 386 processor.
(paraphrasing) "No one needs that kind of speed at home. This is strictly for business servers."
Speed it up, and they will come.
How did this even make in on /.
Should be enough for anyone"
-- Bill Gates
This is like saying back in the dialup days "who needs speeds faster than a T1? It's not like the text is going to get read any faster"
Going to faster throughput makes other things possible that previously weren't.
I don't see the Koreans or the Swedes giving up their fast-as-shit-through-a-goose internet connections because "they don't need it."
--
BMO
This is the most pathetic excuse for an article ever posted by Slashdot. It's complete bunk.
Gigabit NICs are standard equipment. Just because a single machine can't saturate the link due to other IO bottlenecks doesn't make the technology premature or useless. It just means you've got a really, really crappy laptop.
I do not fail; I succeed at finding out what does not work.
This ignores the two principal problems... Latency, and MTU.
Gigabit ethernet connections can use 9KB MTU, My 100Mbps link between Montreal and California can only get 12Mbps. Owing to the 75ms latency.
This is why CDN's (like Amazon, Microsoft, Akamai, etc) are necessary until the endpoints have these high speed connections.
If ISPs offer these at reasonable prices, they will probably add a data cap--like Comcast's 250 GB data cap--much like the new 4th Generation mobile networks, which have a 2GB, 5GB, or 10 GB data cap.
'when for today’s applications, a cable modem offering 12-14 Mbps down will do just fine?'
So we can get better applications. So Netflix can stream without butchering the content like it currently does. Because you really have to worry about multiple users and aggregation. You can really see this with GoToMeeting and WebEx: I don't care what their service claims are, every time we have more than a couple people on a meeting the voice and video are crap.
How about the up being much more constrained than the down? That's not a problem for streaming video in, but is for other applications.
'You don't need more than 12-14 Mbps down because that's all we have today' is a blisteringly dumb argument.
I'm not actually sure theres been any computers for a better part of the last 7 or so years sold without Gigabit Ethernet. Routers and Switches are another story mind you. And it just so happens that IPv6 is a repeat story of that.
"It's very interesting, but I cannot foresee any practical application"
etc etc blah blah blah.
... broadband and on and on.
My organization is on the verge of needing to move our equipment to 10gig soon, because the 1gig network is starting to become a serious bottleneck...
Gee... what third world country or year is the OP posting from??? My workstation, built in 2007, supports GigE and is capable of speeds nearing the theoretical limit, just as it can on a 100Mbps link. I also have several of my servers connected on the same GigE VLAN. As for 10Gig... my old employer had racks and racks of servers which we tested and found to be able to use a significant portion of a 10Gig link. Of course, these systems were using NICs which were at the time (about 2 years ago) running about $700 ea., and connected to switches running about $20K each... but when you build a super computing cluster (to which we had a dedicated 10Gbps connection to the NLR)... you don't skimp on your infrastructure.
Helping build UN*X and the Internet since 1981.
The article is really talking about WANs/ISPs, not networks in general. I don't blame the submitter, though, because even though you can glork from context that this is what it's talking about, the author kept generally referring to "networks" (with a few exceptions).
My first guess is that this isn't an error, but rather someone pushing the religion of "the cloud," where it's presumed peoples' own computer don't do anything but download stuff from outside their home. Talking about last-mile networks and LANs as though they are the same thing, would fit that agenda.
Once you get away from that and back to the real world, though, the idea there are no applications for gigabit networks is a laugh. The lowly "cp" command is one application that comes to mind... I can tell you right now that 802.11n is easily three orders of magnitude higher than the limit of human perception, where I sometimes think "Hurry up" (not that I expect much from wireless -- did I ever mention that I hate my home's LAN? Stupid solid walls!).
I have heard this argument so many times before, but it is just stupid to say: You can only use 400MBit, so better keep your 100MBit instead of getting that full GBit, as you would not be able to use it fully. It might not be 10 times faster, but it still is 4 times faster, which might well be worth the price to some.
Speed is not important, latency is. And even more so are the current problems with buffers.
Here is a presentation on the subject:
http://www.youtube.com/watch?v=qbIozKVz73g
Here is the website which deals with the problem and is trying to fix any problems in Linux (drivers and TCP/IP stack):
http://www.bufferbloat.net/
New things are always on the horizon
We made some really high bandwidth instruments that could saturate a GigE link with UDP packets for less overhead and what we found were two serious problems, mainly some network cards would just start randomly losing packets all over the place when you exceeded a few hundred megabits and secondly, Windows just sometimes decides it is busy and throws the packets away, even after they have entered the system and been logged by wireshark.
The solution, use linux, much more reliable networking, no packet loss on the same machines. Linux however is not acceptable to the customers so must try to work around Windows's random bouts of laziness :)
Obviously, you plug the Gigabit Ethernet into a router and serve multiple computers with Wi-Fi and Ethernet. Then you can run Netflix all day in the living room and still have fast access from other systems.
And all Macs have had Gigabit Ethernet since the turn of the century, with the exception of older MacBook Air models that don't have Thunderbolt. That is a lot of data heavy users, video people and so on.
And any machine with Thunderbolt or PCI-Express has a faster connection than Gigabit, so the idea that Gigabit is too big for today's computers is not right.
For once can we not get the damn network in place and let everything else fall in line later? It'll make the transition much nicer than waiting until everything else is in place and waiting for the network bottle neck to go away.
The title of the article is misleading. It sounded at first, as if it meant the internal network at a company, and not an ISP connection.
Where I work, it was a real struggle to deploy 1 gig internal network switches, and cards. Everyone thought, before I arrived, their mixed network of 10 and 100 was good enough for them. The switches were a mix of old HP 1998 made switches, and home d-link switches, which connected everyone to Accpac, File and Print Servers, and other services.
After deploying new HP switches and upgrade desktops to 1 gb, most complain when connecting using wireless at 56 now.
However to this day, it is still difficult to convince them to spend the extra $50 to buy new desktops with a 10/100/1000 instead of 10/100.
Hope no one at work reals this article title and think the same as I did.
@"Jasper is CEO of an ISP"
That's the key to this biased article. Jasper doesn't want people to have faster connections as it'll cost Jasper more profit. The guy is just trying to manipulate people into thinking they don't need more from him.
Like the old saying, "follow the money".
kidding me! Verizon is getting away with calling its hobbled '4G' network high speed, and now freakin' tweakers are touting gigabit networks?
HAHAHAHAHAHAHAHAHAHAHAHA!
joe six-pack is still on DSL or [ackpht!] cable!
I wholeheartedly agree. The whole point of Gigabit is that one computer can't saturate it. You don't want any one transfer saturating a network.
While a full gigabit on my notebook would be a hog, not just a waste, something more than 100 mbps wouldn't be bad at all, and there aren't many standards in between 1 gbps and 100 mbps.
And, most important, even though one notebook doesn't need a full gigabit, I want the whole pipe coming to my house. Because I want to use my notebook while someone else is using theirs and we're both streaming HD video while the TV is streaming HD video and we're doing all this HTML5ish interactive crap in other windows.
When the choice is "gigabit" that's *only* 420MB of 10/100 that's maybe 70...
YEAH! I'll go for the 420MB connection saying it is gigabit.
When I'm pushing gigs of files over my network I want it to happen as quickly as possible so the network resource isn't choked for hours at a time.
But NOOO!
WE DON'T NEED NO STEENKIN' GIGABIT!
*Cue everything in the omniverse with both a face and a palm facepalming at the same exact instant.*
Chas - The one, the only.
THANK GOD!!!
What I find most telling about the article is that they tested the connection using a single consumer grade article.
I'm techy enough to be on /., but I don't spend too much time at home living in a high-tech wonderland. However, on the network at home we have the following devices:
5 iPhones (view 2 wireless APs)
4 laptops
1 desktop
2 XBoxes with live
Wii
Roku box
AppleTV
All those devices stream media from the internet, play games on the internet, etc. That one single laptop can't handle the connection is fine, but as a household between myself, my wife, and kids, we do cumulatively use a lot of media. Have kept our unlimited data plans for the iPhone, my son has peaked at 8Gb in one month of data usage over 3G. That's on a single phone.
Between video chat, games, working remotely and streaming a desktop session, netflix, hulu, amazon video, etc there is a huge amount of bandwidth required at home. I find that Verizon FiOS does a great job. I can't stand the company, but the product is top notch.
Yes, I'd love more bandwidth at home, just treating a home network connection as a single device is rediculous. I want everything in the house to be networked, and controllable.
having one computer not saturate a link is a good thing. i'm not sure how many networks OP uses with 1 computer, but I with he would all the time, then we wouldn't have to listen to stupid diatribe.
never need more than 640k mentality is idiotic.
Or put it another way, i can get over 4x the throughput, and future proof my network by going gigabit. It's a no-brainer. Particularly if i want to connect to something over iscsi.
I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
You plugged a gigabit connection into a netbook and only got 400Mbps so, nobody needs gigabit? Don't be stupid. File transfers on my LAN typically run 800Mbps, per host. Sure, I really have to work hard to get more than that between two directly connected hosts, but my LAN has lots of hosts. Lots! Servers are often pumping aggregate speeds of 2 or 3 Gbps. With only a few hosts and some premeditation, I can make a 10Gbps switch-trunk groan under the load.
Please, re title the article to say why you don't need gigabit ethernet. But, I need it and could use 10Gbps ethernet in my servers too, but it's frickin expensive!
My old notebook - a Tecra A8 - had a GigE ethernet connection. Many times when I wanted to move a few gig of data, I would skip the Wifi and connect directly to the ethernet. Linux never had a problem reaching high transfer speeds.
It finally came time to upgrade to a new laptop, and to my horror I discovered that many of the manufacturers aren't including GigE any more, because they seem to think Wireless N is enough. Under ideal conditions, I still have to wait a lot longer to transfer large files onto my new laptop than I did with my old.
My whole computing life I've been used to watching things get faster/smaller/cheaper, etc... this is probably the first time I've had to suffer a downgrade because of an upgrade.
Of course the summary leaves out the part of the conversation where bandwidth is also a measure of well, bandwidth. Just because one single individual device can only get ~500MB does not mean that GbE is worthless. What if there are two laptops sharing that connection? It will be tapped out. Put another one on there and all of a sudden a one gig pipe is not big enough.
How stupid are people, really?
Applications will improve when available infrastructure makes them viable. To not upgrade networks when the technology becomes feasible is like saying "what we're doing now with the Internet is all we'll ever need to do and all anyone will ever need". Many of the applications we use today such as streaming HD video, video conferencing and content rich online gaming were made possible by the standardisation of multi-megabit broadband connections. I remember a ridiculous argument made in an Australian computer magazine in the early days of ADSL roll-out that dial-up would never go away because it was a "standard" and our applications had been designed for it. Skype video call over dial-up anyone?
Wrong, wrong, wrong. If you don't want at least a gigabit Internet connection and a 10gig LAN, you're probably a grandma reading this on an iPad.
"When information is power, privacy is freedom" - Jah-Wren Ryel
Even the mightest of internet core routers can only handle tens of ten-g connections, meaning that ultimately the real 'bandwidth limit' of the internet is not going to be at the end points but in the core. So if we rapidly scale the edges, the core will continue having a hard time catching up, rendering 'gigabit to every home' a useless and expensive gesture that will not deliver the full benefit for quite a long time.
The second problem is the one of zombie botnets. When you give the average ignorant end user huge amounts of upload, all you're really doing is arming the bad guys with bigger and bigger weapons with with to launch ddos's. We'e already seen coordinated Ddos's reaching up into the tens of gigabits range; and that is with the average end user having (pulled out of my ass as an ISP figure) say 1mbps or less of cir upstream. What happens when that average becomes 10mbps or more? By limiting upload speeds we are lifting the bar and making it more difficult (eg: requiring more bots) to create such swells. The net doesn't have much good coordination with respect to attacks and so forth and so arming ignorant end users with significant upload speeds will just amplify an already difficult to deal with problem.
I feel strongly that additional protocols are going to be needed in order to give service providers the tools necessary to identify, report, and coordinate responses to DoS related activities. I'd almost even support the idea of new routing protocols to allow victim networks to send signed routing updates saying 'don't forward traffic from these prefixes to us' which can be handled by border routers, to keep dos traffic from leaving the originating domain. ... if we just kick the users of the net, everything would be fine!
Back in the 90s, when I got a 28.8 kbps modem, I felt it was slow, and would really like a faster internet connection. When I used a 30 kByte/sec in the afternoon, I felt 'This is fast enough for me.' 10 years later, I got a budget 768 kbps internet connection, down from the high speed T3, I had been using prior to that. I did not feel the desire to pay more for basic internet and video. With tabbed browsing, I can read a web page, while another loads. 768 kpbs internet connection is faster than my brain, so I am satisfied. I just want a cheaper 768 kpbs connection.
I am talking about solely from my house. I know I can always go some place with a real internet connection, should the need arises.
It's nice to see that kind of speed being sold to the public. Really. And I would be one of those who would pay for it (never mind the benefit, just being able to have it is enough for me -- but that's beside the point). However, once you really look into it, it would be better to have the following: a) Decreased latency. This also goes for the need to be able to cope with everyone maxing out their connections and the amount of packets dropped due to overflow (too many packets too fast) be around 0.00.............1% b) Increased Upload Speeds. This is one that bugs me the most. While downloading works great (you only need to tell the other end "got it"), when you have to send information, or there are too many people using internet, even if you have the awesome speed of 1Gbit DOWN, you won't enjoy it. Fix those two points, and THEN we can talk about increasing what you offer to customers. Oh yeah, don't forget the hardware to deal with it.
I don't care if I'm wrong. I only care about everyone obtaining something from the discussion.
All I know is, working at home, if I -could- get a 1Gb connection here where I live, even at the $300 a month Chattanooga is charging, I would do so in a heartbeat.
If you have sufficient bandwidth
Except one is unlikely to have sufficient bandwidth. Home users MUST NOT exceed either of two data rates: the advertised "burst" rate, measured in megabits per second, or the average rate buried in fine print, measured in gigabytes per month. Both must be improved, and from what I've seen, home users in the United States will continue to have insufficient bandwidth for the foreseeable future.
The current mentality of hoarding everything on LAN connected disk is simply due to the fact that we have insufficient reliable bandwidth available.
That and it's a lot harder for a publisher to "recall" a work stored on the end user's machine. Cases of this happening (such as Nineteen Eighty-Four on Amazon Kindle) are newsworthy.
Thanks for the heads-up. I guess I'll rip out all my gigabit infrastructure and go back to cloning our computer labs at 100Mbps (shudder).
Should we go back to Single Data Rate SDRAM while we're at it?
Please stop being so god damn stupid, you are seriously worse than Kotaku now.
I have a computer, my wife has a computer. My daughter has a computer. If they all slurp .4 gig per sec, that's more than a gig.
Yes, the chances are we won't all be needing as much as we can possibly get at the same time, but .1g because we can only do .4g? Who's the brainiac arguing for this?
A. Computers keep getting faster, and all three machines will be able to slurp 1g in just a few years. Telcos should wait until it's a painful need before acting? (Oh, wait, that's been the model since telcos got into the ISP business: do nothing, get kickbacks from the government for doing it, screw the poor and lie. Skip this point.)
B. When we do need it at the same time, we're annoyed. Better to have more capacity than too little.
C. It seems like the alternative to 1g is not 400m, it's 100m. So we should pick
You are incorrect by default.
I used to think this this limit existed - for home use, I hit a wall at full HD video streaming - you'ld only use ~30Mb. Some mention torrent packages - but the actual utilization of this package would still hit the same limit, making the download process redundant. However, if distributed computing (REAL cloud computing - so memory, processors, applications and data are essentially 'out there') becomes real, 1Gbps would get very slow very fast. For now, just give me a game that utilizes realtime streaming of application binaries, and high quality assets over the Net and I'll be happy. Other than games, I'm at a loss as to what these monstrous home applications could be. Anyone?
But yeah, I'm a 'wait and see' on Thunderbolt.
As long as we have these pesky bandwidth caps, there is no need or point for gigabit to the net.
Sig for hire.
Who has a network with two computers (server and one client) on it? Maybe one computer doesn't need 1000Base-T, but put 5 computers on the LAN and suddenly we need 1000Base-T. Imagine that. What moron wrote this article?
It looks like you didn't comprehend the article one bit if you read it at all. Jasper was only quoted a couple times and his only point was that there is a cat and mouse game of networks not wanting to grow without apps that force them to and app developers not wanting to write apps that need high bandwidth because it's not available to most people.
The author, Stacey Higginbotham, is the one that has a completely flawed argument about gigabit costing too much. She opens up by saying in the first paragraph that gigabit costs $300/m and 100mbit costs $120/m but then in the fourth paragraph she states that Sonic.net (Jasper's ISP) offers them for $70/m and $40/m respectively which also includes VoIP. AT&T was charging us twice as much for a 3mbit/384kbit line and local POTS than Sonic.net charges for a 5.5mbit/1mbit line (we're ~10k feet from the office, if we were closer then it would be 20mbit/1mbit) with nationwide long distance POTS. Sonic.net is a great ISP with wonderful tech support and all my friends have switched to it. They've all gotten similar speed increases and cost reductions.
Cable will do 250Mbps just fine. DSL currently maxes out around 50Mbps, so it does require fiber.
I did a realtime demo of data transfer last month between Caltech (US) and IUCAA (Pune, India) at about 550 Mbps using the 1 Gbps APAN network. Astronomy needs it today and will need much larger ones in a few years to follow quickly fading transients that need to be followed up in a variety of ways and to determine which ones are indeed worth following using archival data in real time.
http://apan.net/meetings/India2011/Session/Slides/OpeningPlenary/Abstract3.pdf
An ordinary PCI interface cannot handle GB rates. You need at least a PCIe x1.
There is a conspicuous lack of affordable PCIe wired interfaces available. Directly onboard chips are feasable. But only if properly designed. A laptop may have other limiting factors.
420 Mbps from Sebatopol?
Maybe saturating a USB 2 Gigabit NIC (480Mb?)
The laptop couldn't handle a 52Mb/s Hard disk equiv?
That's pretty slow.
How can laptop be expected to use 280MB/s flash drives, when they
can't even handle 52MB/s? (One would presume the same I/O offload
on network chips as we have on disk drives being available if the speed were there. -- so to claim it's about the HW, is really STUPID!
My desktop can easily handle 600-800MB/s in a hard disk -- so...for bits/sec, I could benefit, AT LEAST up to 5-6Gbps... likely more...
My home network is crying out for 10Gbps already!...get a home server -- put a few T's on it, and see how fast that Xfers over a 1Gbit network...Not real fast. 125MBps is max I've measured and that was on writes. (where I can have multiple writes outstanding so RTT doesn't hurt as much).
So I'd guess that someone is trying to come up with reasons why they should spend money in infrastructure -- better to throw on caps and claim they don't have sufficient bandwidth for those higher BW users....
The kilobit telephone network installed by Bell over the last 100 years has been quite useful and still is. Good old voice can transmit a lot of information. If they pull fiber to everything that has copper running into it, which they likely will (if they can find a reliable way to splice it that any old dunce can do), we will likely find ways to squeeze a lot more than GB on it. Multimode fiber certainly offers a lot that copper can't do, but the problem is the endpoints are very expensive due to different lasers and wavelengths and stuff. That's where they need to focus on, and of course pulling more fiber everywhere everyday. And maybe lighting up the miles of dark fiber. That will probably take a government purchase or lease though, because they are way to future proofed to drop costs right now. They could drop costs and sell the whole bundle but that won't make them any more $$ than selling OC12's for 10K and having that much less to deal with.
I suppose most of you have not read the article... The article deals with 1Gbit Internet Broadband, not 1Gbit LAN.
While it may in fact be true that most of the current installed base of computers don't have gigabit ports, this is changing. Most new motherboards have had gigabit ports on them for several years now.
The statement is not accurate. Any modern computer can handle full gigabit speed with ease over ethernet cable. If there's a limiting factor, it's in the peripheral network equipment.
What the article is missing, is that you can manage reasonably well at 10 or 20mbit. But jumping to 100Mbit as I have done means a huge improvement on a few applications. Things you'd "just not do" because they would take too much time.
Similarly the network companies are aching to be able to sell you stuff that the network can't handle today. A video rental store just around the corner can rent you a movie when you're "in the mood for a movie at home". But you have to go out and walk/drive to the shop. Wouldn't they sell more if you could rent the movie from your couch? Sure! And you can do that already. But then it takes planning. How many people plan badly?
And another thing. Just looking at the traffic numbers, you can extrapolate when > 20Mbit is necessary. Telcos have to start planning for that and actually invest for that way ahead of time. So it's good that telecom companies are looking ahead and investing ahead of time. In the mean while some advanced users will be happy to be connected to the internet at gigabit speeds TODAY.
Common situations where you'll easily saturate a number of these circuits: -If I want to address an iSCSI NAS from across town practically without shelling out enterprise dollars(to be read 10's of thousands), and I do...I could use a Gigabit connection. - When you have 20, 50 100 Wi-fi cards,each only access at ~54Kb/s, you need tech to deal with the higher aggregate number of transactions while managing large packets and there's a great deal of data flowing. Common situations where you will not: -Posting to Slashdot while downloading porn.... I feel I'm entitled to unimaginable bandwidth, and I'm so disappointed...I mean, I don't want it to travel to me from somewhere, I want to step into it and go there.... but that's just how I roll.....but I need much bandwidth...so I'm just rolling slow....
420Mbps is still four times faster than 100Mbps... How is that not worth it?
You mean the hardware or the Micro$oft software running on it? No, I will not RTFA when the summary sucks ass this much.
Oh Rob, I miss you dearly!
Holy shit... as a developer of broadcast television infrastructure equipment which is used throughout the world, though predominantly in Europe, I thought I was clued in quite a bit on the back-assward methods used within the industry for transmitting TV and Internet signals.
This was a total shocker for me though. I'm a huge fan of using wavelength multiplexing within fiber. Especially when the fiber in my house is a single fiber as opposed to pairs which are much harder to make look pretty in a house. However this is one of the funniest things I ever heard of.
It's taken a really long time for the industry to finally come up with a less than insanely shitty method of using coax cabling for digital media access. Oddly enough, the cable companies have more or less completely rebuild their coax backbones to make it happen... what makes it odd is that they wanted to keep the coax to avoid having to lay new cable.. haha wow that worked well.
Now, it appears that Verizon has decided to transmit the entire cable multiplex over a single wavelength, therefore allowing them to a) guarantee their bandwidth usage even if it's insanely high, b) decrease hackability of TV fiber as it is on a not so common wavelength and therefore difficult for consumers/hackers to get receivers for it. c) run less expensive multiplexers they wouldn't require conditional IP multicasting at the switches. d) decrease the cost of maintaining a huge TCP/IP network of devices as it would be possible to remove the IP layer altogether and use a more reliable ATM style layer.
This design so fantastically screws consumers into buying/leasing equipment exclusively from Verizon that it damn near guarantees Verizon a minimum of $30 a month extra per average household just in equipment rentals. And what's best is, they can claim "Sure, we support using third party hardware with our system Mr. FCC, but there's no law that says we have to help anyone make equipment that works with out network is there? But if anyone ever does... sure, we'll support them".
The only true benefit of this design to the consumer is that it would be possible to make a fiber to DVB-C converter that would theoretically make it possible for a TV to receive the signal using the digital coax connection within the TV.
I am SOOOOO glad I don't live in the states anymore... this stuff would infuriate me... it's bad enough I had to make an FPGA for brute forcing DVB-CAS in order to cut my power usage in the house by 100watts (24/7 since the shutting off the set top box from the fiber company requires a 3-5 minute startup time). Now I use an FPGA which consumes 5watts to crack the keys and shared them out with the rest of the house. Saves me a fortune. The FIOS thing would drive me nuts... oh there's the additional bitch about FIOS which is that it's DOG SLOW!!!!
History is littered with comments of this ilk, but made by people whose opinion should be considered. ./ readers you should hang your head in shame.
"I think there is a world market for maybe five computers."
"There is no reason anyone would want a computer in their home."
This commentary is made by a journo who's only job is to make lots of people go to this web site.
Game/Set/Match to journo,
ps a decade ago I had access to a network with multi-gig connectivity to the Internet as I had a job with a research organisation. It's the only network that I've work on where I actively tuned the TCP stack and optimised disks to cope with the download speeds, it was fantastic. All I can say is bring it on.
that's right. 6 years ago, I was using a gigabit network at home to stream data to a disk at the full speed that the disk could handle. 80 MB/s or so (for bursts, less for sustained transfer).
And yet here you are claiming that computers these days can't handle it. What a joke.
What's more, this story is presumably incited by the story a week or more ago about new advances in light detectors which will supposedly double 'internet speeds' to 10 gigabit.
What you and the other clueless failed to realise is that that increase is for ONE FIBER LINK (single directional, probably) and that internet backbone links are made up of many of these.
The bandwidth out of the UK is already 1.1 Tbps or something like that.
Grow a brain and do some research before posting ridiculous articles like this one.
All I can say is, I cant tell if trolling or serious. Sure my one laptop on wifi (lets not even go to what it can pull when I put it on the good ol' copper) cant saturate my gigabit, but my laptop, TV, Xbox360, PS3, Wii, Desktop, Tweeting toaster, self ordering fridge, etc together will easily saturate that. As we look into the future more and more is going to be interconnected and on the network and eventually on the internet, so sure one system can pull 480mbps but what if i have 50 systems in the place that all use 480mbps?
Its thinking like what comes from this article that has the US so far behind the rest of the world in internet speed, bandwidth, and cost...
Everyone seems to be missing the point here. Laptops are generally configured for battery optimization over performance. This means if the laptop has a spinning disk ( very likely) it will likely be a 5400 rpm drive. These drives are not well know for being a performance super start. The NIC in most laptops is also usually configured for a powersavings mode as well that will result in slower network utilization as well. Then we look at the processors and ram which also are usually configured for power saving mode. Things get even worse if you dont have the laptop plugged into an outlet as they go into an even more restrictive power savings mode. All that being said I would gladly pay $75/mo for gig-e wireline speeds with no cap. If comcast ever offers this people would be going over the bloody cap in the first 45minutes of the month :(
it's a lot harder for a publisher to "recall" a work stored on the end user's machine. Cases of this happening (such as Nineteen Eighty-Four on Amazon Kindle) are newsworthy.
Yes, sure, a publisher can remove access, but thats a separate issue really - they can already do that via DRM embedded into the content.
But there's a lot more bad press when the publisher terminates access to a work that users have already downloaded than when when the publisher terminates streaming access to a work. Take the case of the Kindle and the various WMA music stores that have shut down their DRM servers. It's comparatively no big deal when a work is no longer made available for streaming, but if termination of end users' license to view a downloaded work were to become a regular occurrence, it could undermine end users' faith in a particular technology.
Over the last year have been upgrading everyone I know to Gigabit Ethernet since many home systems are implementing NAS for their data storage needs - and it is barely useful since transferring media files sucks bandwidth. If 10Ge was cheaper I would be pushing that out. Anyone who says we don't need speed is an idiot and should be dismissed as such since they have no clue about even the simple process of moving data around a network, and 1Ge is barely useful when dealing with a home network, for business it sucks rocks and requires segmenting the network with duplicate file storage so that people can get somewhat reasonable response times.
I already have TV through fiber-based internet. No cable, no satellite. But gigabit is not needed for this either.
Looking at the adoption rates of Internet solutions and any other technology you have a Field of Dreams situation of "If you build it they will come".
While laptops may not be able to take advantage of a 1Gb connection two could. 3 really could and looking at the roll out of computers that is they type of household we are looking at. Especially as more video content is delivered over the internet. At some point the cable companies will figure out a better way to deliver their content over an IP based network and that Gb connection will not seem like so much if you are trying to watch multiple HD channels in different rooms while also trying to download the latest video that your aunt Nelly shot of her dogs on her 4k camera. Letting the connection companies give the bare minimum with old and outdated technology will just result in an old infrastruture that will not be able to be upgraded.
In rural areas where the providers didn't want to spend the money on fiber backhauls they were stuck with a cap on their maximum supported speeds and just tried to ride it out by laying more copper to get them by. Unfortunately they spent that money on a technology that wouldn't scale out. Now you have many areas of the country that have broadband available but that broadband is only 512k or 786k because they can get away with it and they don't have the backhauls to provide more than that to all the customers. They also are running over some crappy old copper that gets flakey when they try to push 1mb dsl over it.
By the time its even modestly deployed it'll be 10 years after we need gigabit networks. Start building it now so we're only marginally behind the curve instead of decades.
I just have to clarify, that what we call a "Gigabit" network has 1 Gbps as a theoretical limit on the physical bridge between two nodes. That is to say between your computer and the nearest router with more than one connection. The rest of the network will impose slower speeds on you. I once had my laptop, with a gigabit ethernet port, tethered directly into a fiber backbone at a telco site. I still couldn't even get close to the theoretical limits of that trunk when I was the only traffic on the line because I was downloading a file and the server at the other end limited ftp bandwidth.
Theory is never practical.
If we didn't have bandwidth caps how would our poor broke isp/video content providers punish us for using a competing service?
Because there is only one computer per household. That computer only has one network accessing application. That application only makes one request. And! Only one response returns to that one computer while that one person waits patiently until it is finished before ever making another request.[/snark]
Having to work for a living is the root of all evil.
Yeah, just try telling my company (large hospital) that we should be transferring our nightly, multi-terabyte backups at 100Mbps........
From your view point we shouldn't build anything till we absolutely need it... This thought process is completely idiotic. With that same thought process there would have never been Cable connections. Look at computers back in the late 80's and early 90's. The speed of those machines could never handle today's internet speeds. If higher connection speeds are provided I personally as a developer along with many others will be able to push bigger and better innovations out to users. So if you are wanting to halt progress then by all means... keep thinking only about the now and not 5 years down the line.
When I was working on large farms of big iron (SGI, Sun, IBM systems) we regularly found that GigE over copper couldn't do much better than 400-500 Mbit/s. These were enterprise-class NICs, not cheap-o home gear. Switching to fiber got us much much closer to theoretical max.
When is fiber going to finally be available/affordable for the home market? I think that will make the biggest difference...
Actually some pretty modest hardware will generate very high thruput if tuned properly. The network stack default parameters we use today are remnants of what we did in the the 90s, when 16MB RAM was big, 4Mbit token ring was used, and 100k for a network buffer was a lot.
To increase thruput there are several things which can be done. The first is to increase the window size so that data can flow until the ACK packets get back. The second is to increase the packet size (aka jumbo packets, or MTU). After that you need to allocate enough buffer space to keep the pipe full on the transmit end and prevent buffer overflow on the receiving end. The OS needs to prioritize interrupt handling so the incoming data get handled, it doesn't need a lot of CPU, but it needs it NOW.
Finally, realize that the disk subsystem may become a bottleneck at Gbit speeds, sustained transfer to/from disk may take more than the minimal bargain drive. You don't need super hardware to use that Gbit, but you do need some optimal use of the hardware you have.
Ok, so for two computers directly connected over a gigabit connection trying to transfer a file - yes, it is correct - you cannot max out a gigabit using most computers - the hard drive interface becomes the main limitation, and you'll typically max out around 500 Mbit/second. That's not to say an application that only generates network traffic - that does not need to put it on the hard drive - could not max it out, it very well can since the limitation is the I/O for the hard drive, and not the computer in general (CPU or RAM). This is true even of vintage year 2000 hardware.
However, when talking about multiple computers the systems interleave, and then it can be very easy to max out a gigabit connection even with just two or three computers on the network.
That said, many homes have multiple computers once you start counting: laptops, desktops, iPods, iPhones, Android devices, DVRs, BlueRay, Gaming Systems (e.g. Wii, PlayStation, XBox), and the list goes on.
So yes, Multi-gigabit networks to the home from an ISP are very very relevant and necessary.
Truth is like the sun. You can shut it out for a time, but it ain't goin' away. - Elvis Presley (source: imdb.com)
"It’s a question that analysts posed of Verizon, when they pressed the company that deployed the nation’s largest fiber-to-home network, about take-up rates and boosting subscribers for FiOS."
Verizon:
-Bills you $20,480 per gig when in another country. (Blames foreign ISP charges. Blames their own "outdated and bogged down" infrastructure which is brand new and lag free)
-Bills you $500,000 per gig of texts. (1,000,000 texts at 1KB per text) (Blames their own "outdated and bogged down" infrastructure which is brand new and lag free)
~Verizon rep.
This argument is why we have traffic jams on highways. When the first 4-lane highways were built people assumed we'd never need more than four lanes for any highway. Of course, without the foresight and long-term planning, we're now left with highways that can barely handle current traffic let alone that of the next 10-20 years. And to add on extra lanes now would be impossible in certain areas and extremely costly and disruptive for years in others.
It IS worth our while to start creating networks that are "too fast" for current use. (Remember just 10 years ago when people thought T1 lines were insanely fast and overkill for non-businesses? - They were 1.544 Mbps)
I have the luxury of 100Mb internet in the UK (something my ISP has only just now gotten to grips with, despite having installed it some months past). They are currently paving the way for upgrades up to a maximum of 400Mb.
That's 50MB per second, in real money (shave a little off for luck, of course). That is under 2 minutes to download an HD movie.
Assuming they think about a few 10Gb or 40Gb backbones to support things, the close we get to "click download, drum fingers once, download finished", the less stress everyone has due to network contention. I mean, there's only so many HD films you can possibly download in one go.
There are people who already spend 10, 15 or 24 hours a day downloading as much as they possibly can. Try that on a 400Mb internet connection and you'll be buying one of those new 4TB hard drives *every single day* to keep up with your clearly unnatural pornography collecting habit.
The point is, networks when they get to 1Gb for the average user will suddenly find themselves much more able to cope with overall demand. There will be the initial glut of "wow, look how much I can download" from certain sections of the internet community, but the shine soon wears off that little game.
we are upgrading our "old" 10Gb to 40Gbs and 100Gbs right now.
1Gbs wired to desktop - wireless "802.11N" to everything else.
10Gbs from server to server
40/100Gbs backbone
How can you hit 420 Mbps and still diss Gigabit ? 420 Mbps is greater than 100 Mbps, and we don't have a standard for 0.42 Gigabit networks. Chances are, the throughput was limited by disk I/O or even motherboard bandwidth to the port, as laptops often have to sacrifice performance in favour of size and power conservation.
If I had a gigabit pipe to the internet today, I probably would use about 30% more bandwidth than I currently do - which is already quite a bit - but I would enjoy the experience a lot more since I'd spend much less time waiting for transfers to complete. I could mount remote filesystems over the wire without suffering through long pauses due to FS overhead. I wouldn't need such a beefy laptop to do my work, because I could comfortably remote into my desktop powerhouse at home. I could fling media projects around without having to wait a half-hour per file, fostering real-time collaboration with my teammates.
So we may not need gigabit networks yet, but we'd be fools to turn them down. If the computer industry had been built around needs, we would not have personal computers today, because at the time, nobody needed an Apple II. We wanted them, because they were cool, and then we discovered new uses for the tech.
-Billco, Fnarg.com
Here in Norway, our configuration isn't too much different with regards to the distribution of the signals. The primary difference is that instead of using coax over fiber as you've explained, instead, the signals are multiplexed by using VLANs. This makes it so that the pretty much any common ethernet switch (by common, I mean common infrastructure grade) can be used for distributing the signals. Therefore there's no additional need for being able to broadcast a non-ethernet signal over the fiber.
:)
The drawback as I see it in comparison is that all the television channels belong to individual multicasts and therefore unless you have something similar to the ONT which will join all the multiplexes and buffer and remultiplexes them and then modulates all the new multiplexes as DVB-C. If the provider were to provide MPTS streams on a separate multicast, then this could theoretically be handled by a relatively inexpensive unit... something in the cost level of a cable modem... well possibly even less as demodulating QAM is more expensive than modulating given the substantially less complex clock circuitry.
There are a few companies which have attempted to make DVB-IP to DVB-C gateways, but their systems were not as advanced. What they did was to produce a centralized set top box which would then contain 3 individual DVB-C modulators that would rebroadcast a single channel each. Then using RF based remote controls, the viewer would change their channel at the set top box itself.
Altibox (the biggest fiber provider in Norway) for the moment will do everything possible to guarantee the set top box rental fees since after 4-6 months, those boxes are generating huge revenues for the company. They also want to guarantee that every time you look at the TV guide, you're being bombarded by advertisements for VoD.
I am pretty impressed in the end by the FIOS design after all... I'm not 100% convinced it's the right design, but until things like ethernet switches and jack are more common in a household... meaning that when a house is built, conduits are installed in the walls and a patch panel is present in a centralized location, it might be the best solution possible.
Thanks for the information