100 Gbps Via Ethernet

Doc Ruby writes, "As reported at GigaOM, 'Infinera has bonded 10 parallel 10 Gb/s channels into one logical flow while maintaining packet ordering at the receiver,' bridging 100-Gbps ethernet over 10 10-Gbps optical WAN links. Infinera's press release is here. Further from GigaOM: 'The experimental system was set up between Tampa, Florida and Houston, Texas, and back again. A 100 GbE signal was spliced into ten 10 Gb/s streams using an Infinera-proposed specification for 100GbE across multiple links. The splicing of the signal is based on a packet-reordering algorithm developed at [UC] Santa Cruz. This algorithm preserves packet order even as individual flows are striped across multiple wavelengths.' We're all going to want our share of these 100Gbps networks. The current network retailers, mainly cable and DSL dealers, still haven't brought even 10Mbps to most homes, though they're now bringing fiber to the premises to some rich/lucky customers. Are they laying fiber that will bring them to Tbps, or will that stuff clog the way to getting these speeds ourselves?" Rumors say that what runs over Verizon's FiOS is ATM, to support their aspirations for triple-play.

Comments

100 GBPS

[Intron]

Why should I have to wait 5 seconds to download a movie. Don't they have anything faster?

Re:100 GBPS

[Firehed]

5 seconds at 100Gbps? How high-def is your pr0n?

Re:100 GBPS

[Anonymous Coward]

Your HD can't even keep up with that.

Re:100 GBPS

[UnknowingFool]

Kids! Back in my day we have to wake up at 4am in the morning to collect firewood to send out smoke signals. If you were lucky you would have 8 bits out bit by noon. And those were the good days when some script kiddie didn't try to r00t our fires. You'll take your 100GPS and like it. Now, get off my lawn!

Re:100 GBPS

[`Sean]

Man...and I thought I had it rough having to rely on RFC 2549.

Re:100 GBPS

[kfg]

Smoke? Luxury!

In my day we had to throw bits of shit at each other.

That's right, monkeys invented digital communications. Insert your Al Gore joke here:

KFG

Re:100 GBPS

[Profane MuthaFucka]

I used to work at the phone company as a bit ejaculator. My record was 6 bits in one day. The person you wanted to talk to had to be 6 feet away or less.

Re:100 GBPS

[networkBoy]

Hey in my day you had to be within 4 feet and 4 bits was the max.... wait a second there...

Hey you youngin's are weak! in my day we could whack out one full EBDIC character a day and at a range of 10 feet, so there!

Re:100 GBPS

[lattyware]

He stores it as separate bitmap frames, Duh!

Re:100 GBPS

[Kingrames]

Time to bust out with this knowledge I been learnin' in school this semester.

Gbps != GBPS.

That's 100 GigaBITS per second they're talking about in the article, not 100 GigaBYTES.

And it's likely an expression of bandwidth. It's not going to speed up data transmission unless bandwidth is your problem.

For example, it will still take data a certain amount of time to reach you because there's a limit to how fast it can travel along a wire. (the speed of light in copper, 2.9x10^8 m/s is a good estimate)

Re:100 GBPS

[altstadt]

Um, yeah. You might want to try some engineering courses next semester.

Re:100 GBPS

[ken95357]

Do you loose packets when there are fires for smoke signals in your area? :-)

Rich?

[syrinx]

FIOS is cheaper than cable internet here, if you can get it. Just stick with "lucky", unless you're going to say anyone with broadband is "rich".

I live in a condo, so no luck for me though.

Re:Rich?

[Anonymous Coward]

Imagine a beowulf cluster plugged into...never mind.

Re:Rich?

[kannibal_klown]

Yeh, "rich" isn't a requirement. It's cheaper than Cable around here, and not only "rich" neighborhoods have it. In fact, if you look at some of the places in my state you'd be surprised. Richer areas have no word on when they're getting it, but some rundown areas got it a while ago (both city and in the boonies).

Re:Rich?

[Jim_Maryland]

Recently switched to FIOS myself and I'm saving compared to my Comcast bill.

$117 Comcast (Digital Cable without any of the premium channels, broadband at 7 Mbps (although never actually saw rates near that))
$69 Verizon FIOS (Digital TV without any of the premium channels, broadband at 5 Mbps (actually saw rates exceeding that but generally very close to the advertised speed))

Since Verizon FIOS was available in the area (Maryland), Comcast has been pretty heavily advertising their bundle for new customers where you get the Digital Cable ($33), VoIP ($33), and Broadband ($33). Unfortunately that appears to be a one year deal compared to Verizon FIOS which doesn't appear to have plans to jump up after the initial year (hopefully I didn't miss some small print).

FYI - Using SunRocket as my VoIP with the monthly cost under $17 so the Comcast package isn't an option for me.

Re:Rich?

[PPH]

If its Verizon territory, 'rich' means nothing. Poor might be an advantage.

Verizon recently replaced their old twisted copper cables when my city (Kirkland WA, multi-million dollar waterfront homes, within spitting distance of Bill Gates' house, etc.) widened a road that runs near one of their CO buildings. They replaced it with........more twisted copper.

Re:Rich?

[RaboKrabekian]

I so desperately want to switch to FIOS, or anything besides Cablevision, but it doesn't look like I have any options in my area (Brooklyn). Verizon says no to DSL even, which I find odd, but I can't imagine FIOS is anywhere close to being rolled out if DSL isn't even available.

*sigh*

Re:Rich?

[Jim_Maryland]

I was eager to get off of my local cable provider too. I'm hopeful this competition will drive the prices down (which I can somewhat see with the package Comcast is offering).

I only recently checked to see if DSL was an option in my area and apparently DSL isn't available. Now maybe Verizon disqualified the DSL check based on the fact that I already have FiOS but even checking a neighbors address indicates that DSL isn't available. I don't know if I'd rule out FiOS just on the basis that DSL isn't available from Verizon. You might want to contact your local Verizon office to see if/when FiOS will be available in your neighborhood.

Re:Rich?

[Anonymous Coward]

Does it connect Linux?

Re:Rich?

[ZorinLynx]

When it comes to deploying infrastructure, poor areas make the telcos more money.

Why? Higher housing density. More people in one place means you can serve more customers with the same amount of cable and equipment.

Since rich folks pay the same as poor for the same internet access, telcos would rather deploy to higher density poor neighborhoods first.

Of course this doesn't apply to high density rich neighborhoods like much of Manhattan, but in suburbia it definitely does.

-Z

Re:Rich?

[Ponga]

Rich... poor. Pfft. Does not make a difference if your new fiber going in is subsidized. My grandmother lives in a small community (30K pop.) in BFE Eastern Arizona. The neighborhood is lower-middle class, but all the streets are tore up due to a company installing fiber to the home. Rumor has it that the Latter Day Saints (this is a big Mormon town) are footing a good portion of the bill. Either way, this is the only place I have ever seen fiber being laid to the home, city or not, rich part of town or not.
-Ponga

Re:Rich?

[zeath]

Wouldn't a poor area have a lower adoption rate though? It wouldn't make sense to build infrastructure to a community of 10,000 people where only 1% can afford your service, when you can instead rollout to a community of 1,000 with a 20% signup rate.

Re:Rich?

[Sillygates]

If you lived in japan you would have had 100mbits to the home for the past 5 years: URL:http://www.commsdesign.com/news/market_news/sh owArticle.jhtml?articleID=12805822?

Re:Rich?

[Jim_Maryland]

Alas I do not live in Japan so I must be happy with the local offerings.

BTW, the link doesn't work for me...did the article expire 5 years ago? (sorry...had to do it)

Re:Rich?

[GiMP]

I suspect (since I have no references to cite) that the "poor" often are in their situation due to bad money management. This means that they're more likely to spend their money on "luxury" items. For families with children, on the other hand, it may not be a bad investment for them, the children of such families will obtain a great benefit from the opportunities that such connectivity will provide them. Most parents wouldn't mind struggling a bit financially if it means that their children will have an opportunity to excel, in fact, most parents struggle to provide for their children. Additionally, I wouldn't doubt that there are often subsidiaries to help lay the infractructure in poorer areas. So no, I don't think that broadband infractructure in poorer areas is a bad investment.

Also, most 'richer' areas are populated by older people. I also suspect that many 20-30 year olds live in apartments and new housing developments. Areas with older homes and townhouses are less likely to receive multiple broadband options, as there tend to be an older population in those areas. This is shame if you're young, living in those areas, and it is a shame for your parents and grandparents (and esspecially for you when you stay over).

Re:Rich?

[stinerman]

I suspect (since I have no references to cite) that the "poor" often are in their situation due to bad money management.

The "poor" are that way for a variety of reasons, not just the one you stated. Living one paycheck away from being on the street changes your financial options. I would have been on the street if not for credit cards, which I used to pay bills until I could find another job. I'm still dirt poor because I'm working on paying them back (and then I have student loans coming up when I graduate this spring). What I'm saying is that for every person who spends 1/2 their paycheck on lottery tickets there is one who just got the short end of the stick.

Getting back on topic, I would consider broadband Internet access the best bang for your entertainment buck. If you play your cards right, you can do VOIP via Skype or SIPPhone, thereby killing your telephone bill. If you're open to infringing copyright, you can download pretty much any TV show you wish to see, including overseas shows you couldn't otherwise see using cable.

Re:Rich?

[GiMP]

> The "poor" are that way for a variety of reasons, not just the one
> you stated.

Clearly, that is why I said, "often".

> "I would have been on the street if not for credit cards, which I
> used to pay bills until I could find another job."

This could be considered as bad money management. Although I understand that extenuating circumstances may have led to this path.

> I have student loans coming up when I graduate this spring

So you're a poor college student? Well, thats a completely different thing!

> I would consider broadband Internet access the best bang for your
> entertainment buck

Agreed on all accounts, but additionally, as I said, it also provides opportunities to resolve poverty, so its also a good investment.

Re:Rich?

[mibus]

Not necessarily. The theory I heard is that it's more than balanced out, because in the more affluent areas people can afford to go out more often (eg. theatre, cinema, opera, parties, overseas holidays) whereas the mid-lower end of the scale gets more hours of usefulness out of pay-tv.

Whoa

[JRWR]

Holy Cow, thats alot of data at once, i dont even think my HD can keep up with that, tho playing couter strike will be fun with -90ms lag

Re:Whoa

[cybrzndane]

-90ms hey? They have precognitive online gaming algorithms now?

Re:Whoa

[overkill1024]

Technically that's where all of the anti-cheat algorithms are headed. We could definitely use something like that, but it's cat and mouse.

Bad idea for the home network...

[__aaclcg7560]

Couldn't they come up with a single 100Gb cable specification? The last thing I need is ten cables running from each computer into a monster hub. I shouldn't be turning my home into a cable closet! :P

Re:Bad idea for the home network...

[Anonymous Coward]

Short answer NO
Copper can not support that kind of data rate over any type of distance. The fastest copper connection I've seen is 10GBPS over a single diff pair rated for 9 feet or so. It was meant for rack to rack comm. This was a few years ago but it was still a prototype then.

Re:Bad idea for the home network...

[QuantumRiff]

It is one cable. They split the signal out to the single fiber cable using different wavelenghts. (ie, colors) and run 10 wavelengths over the same cable, at 10Gbs each. For simplicity sake, picture it as 10Gb/s with a red laser, 10Gb/s with a Blue laser, 10Gb/s with a green one... and so on.

Re:Bad idea for the home network...

[Amouth]

cisco has 40GBPS links.. but i think the max lenght is 2m.. they use it for the really high end routers

Re:Bad idea for the home network...

[rufty_tufty]

And when I worked at nortel they had lab kit downstairs that was 4Tbit down a single fibre - the joys of DWDM - used for long distance links so you got the most out of your optical amplifiers/transatlantic links.

Last i heard this stuff was in the field, but it wasn't ethernet (If i remember correctly, a concatonation of a number of STM-64 streams).
For me the news here is that they're using Ethernet (a lan protocol from the 80s) for serious MAN deployments; Not to be sneezed at, but WDM is hardly new technology!

Heh like this will ever see homes

[Durrok]

Hell, why give us even 10MB w/o paying out the arse for it when you have people paying $40/month just for 3MB/512K?

If anything like this ever came out it would probably be shared (obviously) and beyond the standard monthly fee there would be a per MB charge as well.
God I hate USA's internet :|

Re:Heh like this will ever see homes

[aetherworld]

God I hate USA's internet :|

Hrhr, that was fun :D

I'm paying 65 euro (=83US$) a month for 2mbit/512kbit ADSL. And that's with 15GB/month download limit (although fair use, which means they turn it off at 50gb). And that's the cheapest option. And btw, I live in a city -.-

Re:Heh like this will ever see homes

[Durrok]

Wow you are getting raped over there. Where in Europe is this?

Re:Heh like this will ever see homes

[aetherworld]

Austria...

and raped is the correct word -.-

Re:Heh like this will ever see homes

[duguk]

That sounds pretty similar to UK prices. Around £25/mo for 50gb limit/mo. Thats €35 or $50 for around 2mbps with a limit. Oh, plus £10 or $20/15 for phone line from BT.

Unfair, no? Its almost impossible to get unlimited connection - its at least twice that for the same speeds, plus the phone line from BT.

DugUK

Re:Heh like this will ever see homes

[sponga]

Thank god there are more activities in the U.S. to do than be on the internet. Seriously wasn't the major argument about 'USA behind in Broadband' that because of the colder places people lived in the other places in the world so they occupied their times in-door on the internet.

Now I live in Southern California with a huge population and luckily moved my family to a more upscale neighborhood luckily; we are on of the few small communities to get FIOS but I just do not see anybody praising it around here at any school events, when at the kids baseball games or at the adult softball league games. It changes region to region that I have moved to during my construction job; but what I am getting to is the Supply and Demand and how nobody demands more speed.

FIOS around here is now going to try and compete with this new technology with Time Warners deal on the Triple Play Package(phone,broadband,TV) for very cheap($100 month)

Re:Heh like this will ever see homes

[chewedtoothpick]

I pay $80 per month, for 10mbit (down)/ 768 up DSL and Phone (w/ long distance) using Surewest (www.surewest.net) in Roseville, CA. When I am downloading from Fileshack, I have seen 12mbit speeds.

Re:Heh like this will ever see homes

[charlesnw]

Yikes. I pay 27.00 a month for 6 MBPS down/ 768 up. I have AT&T as my provider. I don't use my land line except for DSL and 911 calls. I plan on setting up Asterisk one of these days but generally me and my wife use our cell phones.

Re:Heh like this will ever see homes

[LiquidCoooled]

UK 10mb unlimited for £34.99 (1/2 price first 3 months).
Thats with NTL, not sure if you are still required to get the other services or not (we do anyway).

Its not bad and theres been talk this year of 100mb :D

Re:Heh like this will ever see homes

[Anonymous Coward]

True yes, but thats more than what I said and they Still have limits of 1GB/day.

Plus theres often a bit of talk about ntl being a bit evil...

Anyone know of any good UK Broadband Providers? Freedom2Support used to be good

Re:Heh like this will ever see homes

[LiquidCoooled]

Actually, I have uncapped 10mb service. The minimum 2mb service is still capped however.

As stated on the NTL website: (which has gone for maint since I posted earlier, but google cache is here)


What difference will I see when upgrading from 512K to 2Mb Broadband?

        * Uploads at 256 kbps
        * Nearly 35 times faster than dial up
        * Takes just 4 seconds to download 30 second (1mb) movie trailer / video
        * Takes just 20 seconds to download a 5mb full music single (mp3)
        * Takes just 1min 32secs to download 23mb software application

What difference will I see when upgrading from 2Mb to 4Mb Broadband?

        * Uploads at 384 kbps
        * Nearly 70 times faster than dialup
        * Uncapped - no usage limit
        * Takes just 2 seconds to download 30 second (1mb) movie trailer / video
        * Takes just 10 seconds to download a 5mb full music single (mp3)
        * Takes just 46 seconds to download 23mb software application
        * Free exclusive access to premium content from top internet brands with Broadband Plus

What difference will I see when upgrade from 4Mb to 10Mb Broadband?

        * Uploads at 384 kbps
        * Nearly 180 times faster than dial up
        * Uncapped - no usage limit
        * Takes just 0.8 seconds to download 30 second (1mb) movie trailer / video
        * Takes just 4 seconds to download a 5mb full music single (mp3)
        * Takes just 18.4 seconds to download 23mb software application
        * Free exclusive access to premium content from top internet brands with Broadband Plus

Re:Heh like this will ever see homes

[Anonymous Coward]

Damn it, use the correct units. Megabyte is MB, Megabit is Mb. There is an eight-fold difference.

Re:Heh like this will ever see homes

[WuphonsReach]

My business class DSL with Verizon in Pennsylvania is 1.5Mbit down, 368Kbit up and I'm paying around $140/mo.

But that's a completely unfiltered connection, static IP, no caps, priority support, no hassles about how much bandwidth I use. Which is good, because I'm a digital packrat and I always have something downloading in the background.

Re:Heh like this will ever see homes

[Kheng]

We pay 50AUD/Month for 512/128 Unlimited here in Aus. It's one of the cheapest true unlimited plans. Most plans higher than 512, like 1.5 ... 24Mbit have caps of 20, 40, ... gb and aren't available with Unlimited Download Caps.

Re:Heh like this will ever see homes

[kramulous]

For the last two months, I've been paying $103/month (australian) for 3MB/512Kb (21GB) which has not even been connected. I'd just love to get an ACK at this stage. And I live within 3 kilometres of the second largest CBD in Australia.

Re:Heh like this will ever see homes

[matthew.coulson]

Those articles are 3 years old.

Both the mid and high end services are unlimited both daily and monthly. I'm not sure about the basic.

Re:Heh like this will ever see homes

[Anonymous Coward]

NTL might have a cap, but Telewest Blueyonder (Now owned by NTL) does not. I've been very happy with my Blueyonder cable. Having said that, I recently enquired about a fixed IP, and was told the only way to get one would be to upgrade to NTL Business Broadband (4Mb down, 512Kb up) for £40 a month, and then pay an extra £10 a month for the IP address. I still can't decide if my need for a fixed IP is worth paying twice as much as I currently am (although I do get double the bandwidth as well)

Re:Heh like this will ever see homes

[Anonymous Coward]

iirc, ntl and many other companies have a "fair usage policy" which does not specify a limit, but is down to their discretion. oh, and ntlsuck.co.uk.

Re:Heh like this will ever see homes

[aetherworld]

Oh yeah, phone line MUST be bought here too in order to get ADSL. And it's also an additional US$ 20.

GigE

[bhima]

I'd be happy with something between GigE and 10GigE... seems like they do all of this wonderful shit for the top tiers while the rest of the world gets by with 'fast Ethernet' or GigE at best.

Worse the prices beyond GigE are nothing short of heart stopping.

Re:GigE

[necrogram]

its rather simple.... the guys with the need for 10GigE and faster have the deep pockets. Follow the money

Re:GigE

[anticypher]

The prices of 10GigE switches were outrageous just a year ago, but they've fallen quite a bit as cheap chinese and indian switch makers are getting their production lines cranked up.

I'm seeing about 200 euros per port for a 24 port 10GigE switch, much less for a switch with a few 10GigE ports and 24 10/100/1000 ports.

Give it another year for 10GigE. Where the price savings are now starting to happen is with the real, working 1GigE switches with jumbo packet support, flow control, and non-blocking switching fabric. Those just started hitting about 100 euros for an 8 port switch, and they make a nice home switch for power users.

the AC

Re:GigE

[bhima]

I've seen those. What I'd like is some thing to roll out at my 2nd gig. 20 workstations 7 servers and a network that most people would like to be a little faster than 1GigE we use now... but no one is crazy enough to want 10GigE.

Re:GigE

[Cecil]

Do you have a good network topology? Because that's a lot more important than throwing more bandwidth at it.

Besides if you have a good network topology you should be able to upgrade only the core switches and possibly servers to 10Gig, alleviating your bottleneck without needing to upgrade everything.

Re:GigE

[WuphonsReach]

Aye, consider your network topology.

For servers, you should look at NIC bonding where you use 2+ cables between the server and the switch. Dual-NIC Intel cards are only $170 or so and there are even quad-NIC cards (or you could just install two dual-NIC cards). That will give you 2-4 gigabits per server of bandwidth to share among the workstations.

You may even want to consider NIC bonding for the workstations, but that will require extra wire runs. Or maybe you get an inexpensive 16 port switch (~$240) with trunking and put it into the middle of a cluster of 4-8 workstations. Swipe a few of the links back to the wiring closet and trunk them together with the workstations connecting to the new gigabit switch.

Heck, our network is still only a gigabit core switch with 10/100 hubs hung off of it. I'm still trying to get away from hubs and make to the move to gigabit switches. (But the gigabit switch at the core was a good investment.)

I've seen unmanaged (but with web interfaces, a.k.a. "smart") 48-port gigabit switches for around $1500. I plan on using one of those for the non-critical desktops and it will allow us to retire a pair of 24-port 10/100 hubs.

Reminded me of my modem days

[Anonymous Coward]

It took them this long to do this?

I remember back in the day, there was an attempt to bound multile modems together to get faster throughput. I think they were using 5 54k modems. The down side was that you need 5 phone lines and the ISP on the other end would let you do it.

Re:Reminded me of my modem days

[Amouth]

it was called shotgunning.. it was a hack on how ISDN did it with multi chanels.. it worked aslong as your isp supported it and you had the phone lines.. i know because i used to do 3 of them for a total of a 156k (52k*3) - never got 56k on any of the damn lines.

bonneded DSL was neet too but required all your lines to be F1 pairs and they had to go to the same DSLAM, better to use the F1 pair as either a T1/fractional Frame or PRI - but they charge good money for that.

Re:Reminded me of my modem days

[larytet]

there is a company which does the same for the DSL links (see http://www.megabridge.com/)

the trick is that you can not do this in software, you need hardware to handle the packets. The other problem - rates of different DSL links can vary from 100s kBits/s to single digit Mbits/s. There are more problems.

In case of Infinera they fight jitter at 10Mbits/s. Not a simple "i can do it myself at home" solution.

This is not balance loading - one packet goes here, one packet is sent there. The hardware sends packets bit by bit and balance loading is done on bit by bit basis, not packets. Very likely they have proprietary framing, etc. etc.

Whoa.

[Wrexs0ul]

Buddy, your 62.5 Gigabyte movies are some hardcore HD.

Re:Whoa.

[ebob9]

Well, thats closer, but I think in reality it would be a bit smaller.

So, you did 10 (Gbit/sec) * 8 (bits per byte) * 5 (seconds) = 62.5 Gbytes.

The 10Gb links are Ethernet links. Lets also assume HTTP is the transfer method, just to make it 'easy'. We could use FTP which is UDP but then we'd have to account for the TCP Control connection in the traffic. Heck, Lets even assume nice jumbo frames with a 9000 MTU. Also, lets assume the video is 'optimally' compressed.

Ethernet header = 14 bytes
IP header = 20 bytes
TCP Header = 20 bytes

14+20+20 = 54 bytes out of every 9000 transfered for header information.

On top of this, there will be HTTP headers at the start of the request, but since they are only transfered at the start (not every packet), lets factor them out as miniscule.

So, basically 62.5 is the maximum theoretical data of the circuits. 62.5Gbytes /9000 * 54 = 375Mbytes of packet overhead.

The Maximum possible transfered in 5 seconds would be 62.5Gbytes - 375Mbytes = 62.125Gbytes.

Re:Whoa.

[joshetc]

Where did you get your 12,000MB/s memory from? Thats like 1.8x faster than my dual channel PC2-6400!

Re:Whoa.

[keithpreston]

10 (Gbit/sec) * 8 (bits per byte) * 5 (seconds) = 62.5 Gbytes. This is wrong 10 G Bit/sec / 8 (bits/ byte) * 5 sec = 6.25 Gbytes

Re:Whoa.

[Feyr]

FTP is still TCP. you're probably thinking RTP

Re:Whoa.

[ebob9]

Yup, your right.

I was thinking 10(Gbits / Sec) * (1 byte / 8 bits) * 5 ( sec ) = 62.5 Gbytes

I noticed the units didn't cancel correctly, and didn't think it through.

Re:Whoa.

[Anonymous Coward]

Ebob,

Quite correct. I hadn't factored in overhead, lost packets, or latency.

With your calculations alone though, I tip my hat to your packet powers :-)

-Matt

Re:Whoa.

[Anonymous Coward]

10 Gbit ethernet is 1 GBPS due to 8B/10B coding

Re:Whoa.

[cheater512]

FTP uses TCP not UDP.
I think your thinking of TFTP which uses UDP for things like network booting.

Re:Whoa.

[indigest]

Don't forget the LDPC error correcting code in the 10GbE standard. Every block of 1723 data bits is encoded into 2048 coded bits.

Re:Whoa.

[fozzy1015]

Ethernet header = 14 bytes
IP header = 20 bytes
TCP Header = 20 bytes

An Ethernet header is 18 bytes. Don't forget the 4 byte footer for CRC.

Re:Whoa.

[fozzy1015]

An Ethernet header is 18 bytes. Don't forget the 4 byte footer for CRC

Ok, that sounded lame. I should have just said, 'Don't forget the 4 byte footer for CRC for the Ethernet encapsulation'

Re:Whoa.

[nyri]

So, you did 10 (Gbit/sec) * 8 (bits per byte) * 5 (seconds) = 62.5 Gbytes.

Or maybe he did 100Gbit/sec / 8 bit/byte * 5sec = 62.5 Gbyte.

Sorry to pick your numbers as your point was good but the readability of your comment was, at least for me, quite bad as the formula was not only wrong but misleading.

Re:Whoa.

[monsted]

If you want to be really anal about it, transmission speeds are in SI units, so you'd have to multiply with (1000/1024)^3 :)

62.125 * (1000/1024)^3 = 57.858 GiB

Re:Whoa.

[Jherek Carnelian]

Ethernet header = 14 bytes
IP header = 20 bytes
TCP Header = 20 bytes

There is no TCP header on fragments.
Thus that 20 bytes of TCP header is only once per 64K (maximum TCP packet size).

Clog?

[Nemetroid]

Are they laying fiber that will bring them to Tbps, or will that stuff clog the way to getting these speeds ourselves?

That depends on how many internets they try to send at a time.

soon

[Polly_Morf]

With this new technology I can soon have all porn i the world!

Clogging the tubes

[Anonymous Coward]

or will that stuff clog the way to getting these speeds ourselves?

Well, in the event that it does clog the Internet, you can just use lottery balls to clear the tubes.

Most home routers don't even support 10Mbps WAN

[Anonymous Coward]

The average home user's router gets 5Mbps WAN connection to a modem (cable, dsl) or to the fiber panel that FiOS puts in. Some just now get 10Mbps and even 100Mbps (rare) but my connection is 15Mbps so on my 10Mbps WAN, I lose 5Mbps. (Yay, I can do math.)

You can give us Tbps and we'd still only get at most 100Mbps or even 1Gbps with the current copper Ethernet.

Shows how far behind I am..

[Channard]

... I was happy just to discover I could connect my computers via the electric mains using Homeplugs.

TCP-PR = neat stuff

[neuro.slug]

Here's a link to the paper (PDF) on the packet reordering if you're interested. Being a former banana slug, I was very interested to see this research coming out of UCSC. It makes me feel warm and fuzzy whenever something theory-based is actually implemented.

Re:TCP-PR = neat stuff

[El Torico]

Thanks; I've only taken a cursory look at this, but I see where TCP-PR may be applied to Disruption Tolerant Networks.

Re:TCP-PR = neat stuff

[Anonymous Coward]

The fuzzy part is caused by the purple-haired sinsemilla buds. The warm is the bong water you just spilled.

ZAYANTE

What exactly is new here?

[Anonymous Coward]

What's new here? Isn't this essentially the Ethernet equivalent of packet reordering, which the entire Internet has been doing since day one?

Re:What exactly is new here?

[larytet]

i think this is not packet reordering. Likely load balancing is done on bit by bit basis. They send a packet over all links simultaneously and collect the packet on the other side using heavy hardware trickery.

Yeah.

[Spazntwich]

Ass to mouth is probably a good guess for what runs over Verizon's FIOS pipes.

Re:American internet is worse than third world.

[stupidmonkey]

Europe has the convenience of having small areas to cover. Even Sweden [http://en.wikipedia.org/wiki/Sweden], which is one of the larger countries by area, is just barely bigger than California [http://en.wikipedia.org/wiki/California]. If that was all we had to worry about, I'm sure that ISP's would not be shy to try to blanket the entire country in hi-bandwidth goodness. As it is this is a big ass country, and the cost for building up that type of infrastructure is huge. Of course we're starting to see places like NYC get higher bandwidth coverage, but again, we're back to the case of having a large population in a relatively small geographical area.

Re:American internet is worse than third world.

[PitaBred]

Where do you live? Because Comcast costs $60 for just internet, and that's only around 3mbps if you're lucky.

Certainly welcome in the data centre world

[anticypher]

This is a natural progression of ethernet speeds. 10GigE switches are getting to the price point now that we are installing them everywhere. I even had a 10GigE switch on my home fibre for a week of testing, but slashdot just doesn't load any faster.

All the broadband providers are moving to larger pipes now, with GPON (Gigabit Passive Optical Network) going in everywhere, as consumers are clamoring for more than ADSL2+ speeds (24Mbps down, 2Mbps up) in city centres. I'm designing the back end of a GPON network, where every neighborhood gets 2.5Gig down, 1 Gig up, shared between 16 residences. Of course, there is going to be more than just internet on pipes that big, quadruple play to start, and as new services become available even more bandwidth will be needed. Once you start piling up the 10GigE connections, it will be nice to have a working trunk/etherchannel/bonding solution for those long hauls between data centres.

the AC

Re:Certainly welcome in the data centre world

[D2!R2]

Putting a 10GigE switch in your home won't do anything if your internet connection is still the same. If you want to notice a difference try copying a couple Gig folder from one computer to another one on the network.
And since you seem to know your technical stuff with GPON, do you know if your uprate is time divisioned among the 16 channels or is it a solid 1 Gig upload per channel?
Because I can see a lot of people setting up servers if they can get 1 Gig upload.

Re:Certainly welcome in the data centre world

[Anonymous Coward]

GPON doesn't specify the scheduling method. Most providers will use some sort of dynamic bandwidth allocation to divide the total 1Gbps upstream. So, your uplink speed will depend on everyone else's demand. (And yes, the equipment will be capable of enforcing bandwidth guarantees, minima, and maxima.) If everyone has the same service and the scheduler is fair, you can expect 1/16th of 1Gbps if the PON has 16 users. (More likely 32, or more, if they can stretch the optics.)

Re:Certainly welcome in the data centre world

[Anonymous Coward]

GPON isn't "going in everywhere". As far as I know, there have been around 200,000 lines of GPON installed. GPON-based PowerPoint slides are going in conference rooms everywhere, but nobody's really deploying the equipment in serious volume, certainly not in the USA.

Various carriers in Asia are deploying a lot of EPON (Ethernet PON, standardized as a couple of new clauses in 802.3 for some reason instead of getting its own "dot number"). And by "a lot", I mean millions of lines already in service, not just projected or planned, but bought, paid for, and earning revenue from customers.

There's some BPON going in, particularly with Verizon's FIOS service. But that's 622M down, 155M up, not the same as GPON. And even with GPON, the 2.4G down option isn't all that common in practice.

Re:Certainly welcome in the data centre world

[Yetihehe]

All the broadband providers are moving to larger pipes now

I KNEW it's not a truck!!!

Re:Certainly welcome in the data centre world

[anticypher]

These are European capitals I was talking about. Berlin, Paris, Brussels, Amsterdam. I've talked with a company trying to do a rollout in London, but they are running into a bureaucratic nightmare with Ofcom playing obedient lapdog to moneyed interests. I know there are some upgrades going on in Stockholm and Helsinki where much fibre already exists. Other cities have plans, but it all depends on the local telecoms regulator keeping the incumbents from screwing over the free market.

There are supposed to be over 200k GPON lines lit and generating revenue by the end of this year around Europe. I have serious doubts about hitting that target, mostly because the companies will not hire enough experienced engineers and the marketing campaigns suck. There remains a huge investment in developing the features at the headends, such as TV on Demand, and the CPE boxes are still lagging in features. The sweet spot for pricing is supposed to be around 80 euros per month for a quad package, but the first installs are going for 120 euros or more, and uptake is slow.

The biggest problem right now is the cost of the GPON equipment, the makers just can't get the prices down to where they promised they would be.

the AC

Re:Certainly welcome in the data centre world

[anticypher]

My internet connection is a fibre from my main router to my house. I convinced a supplier to lease me a section of their inter-city fibre that was lying dark, and I paid a crew to pull a fibre from the local pop to my place. I normally have a cisco 3550 on the end of the fibre, and the data centre end is a router with multiple Gig connections to various parts of the internet. I traded my local town for some right-of-way in exchange for the fibre going past the main municipal buildings, so I became the ISP for their IT guys. It works out for all of us, nobody really pays bills in either direction.

For initial tests of the 10GigE equipment, I put a pair on my fibre to see if their promised wan-phy interfaces actually worked. The layer 1 was acceptable and layer 2 worked barely. But with only 4x 1GigE connections further on, there just wasn't any difference moving from 1GigE to 10GigE for home use except for bragging rights. I have a lot of geek friends always trying to out-gadget each other, but nobody has ever topped me in the bandwidth category.

In GPON, the multiplexing depends on the constructor, but typically it is shared between the 16 customers with an ethernet style of contention. Some bandwidth is kept reserved for all the services, internet access is always lowest priority. A few people running popular web servers from their homes might notice each other at peak times, but a few web servers wouldn't hurt. Gamers love GPON because the interleave is so short they see sub milliSecond pings to nearby game servers.

the AC

I'm confused

[Anonymous Coward]

How many Libraries of Congress per fortnight is that?

Yum!!!

[Anonymous Coward]

VR porn!!! I can't wait.

FTP uses UDP?

[yabos]

Not according to the almighty wikipedia
http://en.wikipedia.org/wiki/Ftp

Re:FTP uses UDP?

[ebob9]

Whoops, your right. I got the ftp-data channel mixed up in my head with TFTP (Which uses UDP).

http://en.wikipedia.org/wiki/Trivial_File_Transfer _Protocol

Re:American internet is worse than third world.

[Kyokugenryu]

In the Metro Detroit area, and Comcast offers 8mbps service out here for $60 a month. There's a single competing CableCo, but I live in a small city with a Comcast contract. I refuse to buy anything from Comcast, because they charge far too much for their service. My DirecTV is far better than Comcast's cable service, so the only thing I'm missing out on is the fastest internet service in the area.

Not for grandma

[mypalmike]

'... an Infinera-proposed specification for 100GbE across multiple links...' The current network retailers, mainly cable and DSL dealers, still haven't brought even 10Mbps to most homes.

10Gig+ on the internet is the realm of carriers and huge-volume servers. Cable companies are the customers here. Grandma? Not so much.

Fiber's still the wave of the future

[Manchot]

Let's face it, for long-term benefit, fiber's still the way to go. Though still mostly in the research and development mode, there are companies who can make complete wavelength-division multiplexed optical systems on a chip. Some of them can send and receive 40 Gb/s on 40 different channels. Do the math. That's 1.6 Tb/s per fiber. If you have a bundle of 100 fibers, you're starting to push petabits per second. Also, keep in mind that the main limiting factor for optical data transmission rates is the electrical speed of the transistors at both ends, not the fiber itself. As transistor speeds improve, the maximum data transfer rate per channel will improve. The maximum data transmission rate of copper, on the other hand, is pretty much fixed by the fermionic nature of electrons.

Re:Fiber's still the wave of the future

[Anonymous Coward]

You're right -- fiber is the way to go. As evidenced by the demonstration using DWDM wavelengths for transport.

If the signal was contained in a single wave, I'd be a lot more impressed at their claim of "100GE". What they've provided is more similar to a "10x10 EtherChannel".

Re:Fiber's still the wave of the future

[iWill]

This (Infinera) is the company that makes complete wavelength-division multiplexed optical systems on a chip. They have demonstrated 40 Gbps x 40 channels = 1.6Tbps in development and currently have many 10 Gbps x 10 channel networks deployed by companies such as Level 3 (which first came online in 2004.) Also, the limiting factor in single optical channel transmission is not the transistors at each end. The real problem is the laser modulation at the transmitter as well as the response of the photdetector at the receiver. GaAs or InP transistor circuits at either end are more than capable of handling speeds in excess of 40Gbps if the laser generation/detection ever gets there.

Another one done before...

[Anonymous Coward]

...by a little known free software project.

OT

[Amazing Quantum Man]

Hey Neuro, when'd you graduate? Crown '84 here.

[note to mods - self modded down with "no karma bonus"]

Re:OT

[neuro.slug]

Crown College, 2005, BS in CS (though should have been 2003 if everything went according to plan). Thinking about going back for grad school, as I'm still living in SC.

Rumor?

[Shotgun]

I can guarantee that the FiOS is running over ATM. All the boxes on the houses switch each of the services into seperate ATM VCs. It's the easiest and most logical way to handle it if you want to maintain any sort of QoS guarantees between the different services.

don't you mean

[tpjunkie]

Tubes?

What kind of news is this?

[warrior_s]

Bell Labs did this in March http://www.lucent.com/press/0306/060308.coi.html

Why stop there?

[tringstad]

'Infinera has bonded 10 parallel 10 Gb/s channels into one logical flow while maintaining packet ordering at the receiver,' bridging 100-Gbps ethernet over 10 10-Gbps optical WAN links.

So what's preventing them from taking 10 of these newly created 100GbE channels, applying the same technique, and producing 1TbE?

-Tommy

Re:Why stop there?

[Anonymous Coward]

That isn't technological breakthrough, that is just brute force.

All that means is that instead of 10 GbE cables, there are going to be 100 GbE cables.

Re:Why stop there?

[FooAtWFU]

Well, theoretically, you could do that, but you'd be running into some pretty intense packet-reordering action at the receiver. Heck, they're probably running some pretty intense packet-reordering stuff at the receiver already. You'd probably need more/bigger/faster chips on both ends to do it.

But if you're going for something like that, why bother trying to stack ten 10-way systems instead of just scaling this thing up to one 100-way system?

Re:Why stop there?

[TeknoHog]

It's all about bandwidth, which is not the same thing as date rate. Unfortunately there's a huge marketroid conspiracy trying to teach us otherwise.

Data rate (in bps) is proportional to bandwidth (in Hz). The factor between them depends on the modulation and coding schemes, which in turn are limited by the signal/noise ratio of the medium. Anyway, the system of light sources, fiber, and receivers has a certain limited bandwidth. For example, if you're using visible light from about 400 to 750 THz, you have a bandwidth of 350 THz, assuming the fiber works across the whole spectrum.

For a naive example, imagine that it takes 35 THz of bandwidth to achieve 10 Gbps over a certain fiber. Then you can have a maximum of 10 of these channels using visible light, if you shift the center (carrier) frequencies accordingly (i.e. different colors) to avoid channel overlap.

For another explanation, imagine using a single frequency of light to send information. As soon as you start switching it on and off, you're introducing other frequencies; the spectrum gets wider, the faster you switch (see also: Fourier transform). So, if you try to use zillion distinct colors to get zillion channels, they will start to overlap at some point if you try to send any information.

Re:Why stop there?

[Tesla Tank]

My apology for not RTFA first, but my guess would be the controller they're using to assemble the data at the receiving end cannot process more data. Imagine a controller running at 1 MHz, it cannot process 2 million packets since it's not fast enough to look at each packet. So unfortunately, you can't just linearly scale up the system like you mentioned. This is probably very similar to the problem that we're facing today with optical networks. The amount of data we can pass through the fiber is too much for the switching equipment to handle, so we have to throttle the data.

If the actual article has the real explanation, please enlighten me.

Why is this news?

[cafucu]

Foundry's been doing this for a while now

Re:Why is this news?

[Anonymous Coward]

Foundry goes to 10GbE

This one goes to 11.

Re: not so whoa

[indigoid]

Increasing the bandwidth beyond a surprisingly small figure does not (automatically) improve noticeably the RTT. This is clearly demonstrated in one of the utterly wonderful Stevens books, though I forget which. Most likely one of the three TCP/IP Illustrated volumes.

Ultimately the limiting factors are (a) the transceivers terminating each segment, (b) software, and (c) the speed of light. It sounds like these guys have put their work into (b).

Re:American internet is worse than third world.

[Overzeetop]

Yeah, but Sweden has less than a third of the population of California (20 people/km^2, vs 84 people/km^2), and California is only the 12th densest state. In fact, Sweden is less dense than 32 of the 50 states. So how come we don't have 20MB connections for $15 in those states? Lord knows the big telcos don't bother to serve most of the outlying regions, so we know they already cherry pick the population centers.

It's good old protection. Contracts with towns for exclusive cable rights, and a lack of any meaningful oversight of the telco networks means no competition. And no competition in a corporate setting means high prices. It has nothing to do with population density.

Riiight.

[Kadin2048]

Thank god there are more activities in the U.S. to do than be on the internet.

Yeah ... like eat.

Somehow I think that if we just had so many other things to do, to the point where people just don't care about internet access, because they're just so darned happy to be outside playing softball and everything else, that we wouldn't be one of the most morbidly obese countries on the planet.

I've got another theory: the demand for Internet doesn't exist in the U.S. to the same extent it does in other countries, because people here prefer the passive entertainment of watching television. Which incidentally, they can do while eating.

Re:Riiight.

[sponga]

So even if they had internet I guess they would still be eating in front; either way they do not care and the broadband upgrades is not a high priority.

Thank you for mentioning eating but I get out to meet lots of people who have some of the widest variety of activities to do than any other person I have meet in another country in the world; maybe because of our climate and geographical location.

You would be surprised if you lived in Southern California at how many people go to the gym, run in the morning, hundreds of bike riders on P.C.H. or are out there with me surfing/swimming in the morning; theres stuff to do at night besides stay at home watching T.V. like going out to the bar to meet a girl or going to see a concert.

Yah we have an obese problem but we also have the greatest athletes in the world and you are forgetting we win the most medals in the Olympics; don't turn a blind eye to some of our greatest achievments and try not to single out the poor neighborhoods who do not have time to worry about their health. I will let you be the first to tell the African Americans in downtown L.A. that they cannot wait in those long lines to eat the greasy chicken or tell the Mexicans to start eating salad. (note: I am part Mexican my self)

First ever automobile travel at 500 mph

[dalewright]

OK, so I rented 10 compact cars from Hertz and had each one motor on up to 50 mph.
Tomorrow I might try to get them up to 80 each for an 800 mph demonstration.
Think I can patent this?

Re:First ever automobile travel at 500 mph

[jZnat]

Cars move in a wavelike manner? I did not know that...

Re:First ever automobile travel at 500 mph

[Anonymous Coward]

It's more like saying "I rented ten compact cars and I could suddenly carry ten times the amount of stuff in the same amount of time", isn't it?

Average density is a meaningless figure.

[Kadin2048]

While I'm not disagreeing with you that many municipalities in the U.S. are effectively corrupt, and have entered into exclusivity deals with cable and telcos that are holding back service deployments, I think that the population density figures that you're using are misleading.

Just taking a country's or state's population and dividing it by its area doesn't give much of a meaningful figure of population density. People don't obey the Ideal Gas Law and just spread out evenly over an area. If that was true, then each person in Canada would be sitting in the middle of a miles-wide patch of empty space. People choose to live in high-density areas, and at least in my experience, this trend is stronger in colder climates.

So Sweden might have a low population density, but that doesn't really say anything about how tightly people are "clumped." I'm not sure what kind of measurement you'd need in order to compare that, but it needs to be something that gives you an idea of how much area is inhabited at various densities. Not just urban development, but small towns and villages that are clustered with lots of space in between. I have a feeling that Sweden, if you looked at it from that kind of analysis perspective, would probably be a lot closer to Canada (low overall density but much of the population concentrated) than a suburban-sprawl area like California or most of the East Coast (the "Boston-Atlanta Metro Area" as Gibson once predicted).

The failure of broadband in the U.S. should go principally to local governments and their shortsighted dealmaking, but the geography and urbanization here doesn't help at all, and neither does the way telecommunications are regulated at the Federal level (more corruption/influence, no doubt).

Re:Average density is a meaningless figure.

[Anonymous Coward]

Well, less than a 10th of the total population of Sweden live in the northern half of the country. 1/3 of the total population lives within 100 miles of Stockholm.

Mod these idiots up!

[werewolf1031]

Good God... My kingdom for two mod points! Parent and GP have just made my day, and probably yours too since you read this far. Well done, guys. :)

what's so special (besides the speed)?

[Creepy]

The idea of packet order guarantee and ethernet are pretty much mutually exclusive, and upon further review (reading the fa), it reassembles out-of-order packets pretty much just like TCP. About the only new thing is the out-of-order disassembly and assembly and the overall speed. It still has the same flaws as Ethernet, which is that it really is only about 80-85% efficient - after that you would be better off with a Token Ring or other managed protocol (token rings are excellent for saturated networks but poor for low saturated networks).

Sigh - and the poster seems to think ATM is a good protocol, but ATM is a terrible protocol, especially for data, but even for voice it's mediocre. It was designed for voice conversations over high noise lines with significant data loss (copper) and predominantly used over low noise high speed lines with almost no loss (fiber). Its advantage is standard packet length (53 bytes) and speed. Worst disadvantage - almost 10% overhead (5 bytes of every 53, or ~9.4%). ATM also has no guarantee of sort order or collision avoidance (since it's asynchronous) so in practices it can be really bad. Incidentally, my networking class voted this the worst protocol back in 1996, but expected it to succeed mainly because of telecoms pushing it.

Re:what's so special (besides the speed)?

[Anonymous Coward]

Sigh.

Compare the 10% overhead of ATM on voice verses the packet header overhead on a VoIP call.

Lets assume, say, you are using a codec that has an 8:1 voice compression, so that 64kbps voice stream is now 8kbs per second, or one kilo-byte-per-second. That means each millisecond of voice is represented by one byte of data.

Keep in mind to gather one kilobyte of data, you have to wait 1000 milliseconds to do so -- that's 1 second of voice delay right there. You just cannot have that much delay in a voice call for the voice to sound good. Factoring in network delay and other issues, you might want at most 20-40 milliseconds of delay.

So in ATM land, you can fill up one cell and fire it off, 10% overhead. In VoIP, you have to send out a packet with maybe 40 bytes of data vs. a 40 byte packet header. Talk about your overhead! It just gets worse with IPv6.

This isn't to say ATM is a great protocol, but calling it terrible is a mistake. For a pure voice-only network, it would be superior to an all IP solution.

This is like complaining a motorcyle makes a terrible passenger bus.

Re:what's so special (besides the speed)?

[Creepy]

I didn't mean TCP is a great protocol for voice (it isn't, though it might be for sample and send [uncompressed] data). My problem is with the choice of ATM, which was picked because it matched hardware the telecos were using at the time, not for being a great protocol for voice traffic. It actually is a bad protocol for the switches used by telecos today. The telecos also standardized ATM on their trunk lines for the Internet, even though it's a terrible protocol for data (using 100 motorcycles when a bus would work just as well). ATM also has a packet verification stage (guarantees QoS or "Quality of Service"), which increases the total load. I believe the average was around 17% load both ways, where load meant something like header + send verification + average packet loss at 85% capacity.

    VoIP does not have a QoS guarantee - VoIP uses UDP (Universal Datagram Protocol), not TCP which does not guarantee packet arrival. UDP uses 20 byte packets over IPv4 (TCP is 36 I think). Since UDP does not reply back to the sending machine, it may actually work faster on average, but may also have jitter, especially over long, noisy, or very busy connections.

    One way to counter the empty packet thing that you mentioned is a technique used in cell phones applied to IP - TDMoIP, which bundles a bunch of packets going to the same destination together.

I'm impressed

[Anonymous Coward]

To paraphrase Blue Thunder: "And that, gentlemen, is one hell of a porn-storm in anybody's language!"

Luxtera

[gal1264]

Well, this is cool, but what does it mean? A startup in San Diego (Luxtera) is already sampling both monolithic multi channel xfp modules made entirely from silicon (minus the laser which is indium phosphate), and also makes a silicon dwdm system as well. There have been a couple of news releases lately.

News about DWDM
http://home.businesswire.com/portal/site/google/in dex.jsp?ndmViewId=news_view&newsId=20061017005207& newsLang=en

Darpa Grant Continues
http://www.convergedigest.com/DWDM/DWDMarticle.asp ?ID=19886

FIOS in Maryland

[TubeSteak]

As much as I'd like FiOS, from looking at Verizon's Maryland FiOS page, Verizon isn't rolling out Fibre until the county has granted them a franchise to do TV over FiOS

The most recent article discusses the fact that Verizon sued "Montgomery County asking the court to require the county to negotiate a lawful franchise with the company. Verizon and Montgomery County have agreed to stay Verizon's lawsuit until the county council votes on the agreement. If the agreement is approved, the case will be dismissed."

So, while I don't know if Verizon is bundling the service to consumers, Verizon is certainly bundling the services to the Counties, since Verizon already has whatever permissions it needs for internet and voice services.

Not to mention that it's more than likely Verizon has no intention of investing in a FiOS network for poorer, more rural counties.

Re:FIOS in Maryland

[ScentCone]

So, while I don't know if Verizon is bundling the service to consumers, Verizon is certainly bundling the services to the Counties, since Verizon already has whatever permissions it needs for internet and voice services.

My Mom (in Montgomery County) has FIOS up to her house for data. It's fast. She's using Comcast for cable TV. She called Verizon about switching over to them for 'cable' service so she could bundle that, and her copper land line all together. They said "Sure! we'll have someone out to talk to you about that tomorrow morning!" She asked me to stop by just in case there were going to be "any of those techie issues."

Well, there was. The "issue" was that Verizon's cable service was actually an affiliate sale of DishNetwork (or was it DirecTV?), and there on the lawn was a guy ready to bolt a dish to the side of her house. And it was a 2-year commitment! I'll allow for 10% of Mom not knowing what she was doing on the phone with Verizon, and 90% Extremely Sleazy Customer Service Tactics. Incredible. So, with the guy standing there, screwgun in hand, I called Verizon, who told be that it was "too late" to cancel the "order" but that in 24 months, we could switch over to FIOS TV, which would surely be available by then. We told them that it was also too late for the lawyer's phone to stop ringing, and presto the installer (who spoke no English!) was getting a cell call, and he had the dish packed back up and in his van before we got into our second cup of coffee. Beware the cable/FIOS state of flux in Maryland, especially Montgomery County. Of course, you also have RCN as another option.

Re:FIOS in Maryland

[Jim_Maryland]

I'm in Howard County so this article covers the TV offering:

Howard County approved Verizon's franchise agreement on Jan. 3. The company expects to begin selling FiOS TV in the county by the end of April

I held off switching until I could do a complete switch over to FiOS. I think the package offering from Verizon FiOS only applied to savings with the VoIP offering (might have been a $5-10 difference) but as I stated above, I already have SunRocket and it is costing a lot less than Vonage, Comcast, or Verizon VoIP offerings.

Verizon sued "Montgomery County asking the court to require the county to negotiate a lawful franchise with the company.

When I worked for PG County, I recall that the county had some network deals setup with Comcast to connect the county offices. I wonder if Montgomery has similar deals and if they included anything to keep Comcast as an exclusive cable TV provider.

Re:FIOS in Maryland

[Jim_Maryland]

Earlier in the year when we had inquired about FiOS, we had the same issue. We could get the broadband but we'd have to use DirecTV for our TV service. We just waited it out until all services were available. We did however contact Comcast and ask them to compete on price before we even made the switch. Comcast did lower our rates on the Internet service but they wouldn't budge on the TV side of the bill. Still, we managed to cut the bill by about $15 until we completely dumped Comcast.

Re: not so whoa

[cheater512]

Actually it would decrease pings to quite a degree.
First off this is FAST so the equipment being used is good quality.
Second if a program wants a packet sent it doesnt need to wait as long for the packets ahead of it in the queue to be sent.

This message was posted over a network with 0.117 ms lag. :)

Darth Verizon

[DragonHawk]

"Verizon FIOS which doesn't appear to have plans to jump up after the initial year "

I have some experience dealing with AT&T ^W New England Telephone ^W^W^W NYNEX ^W Bell Atlantic ^W^W Verizon. I can say with assurance that there *is* a plan to increase the rates. You just may not be aware of it yet.

"I am altering the deal. Pray I do not alter it any further."

Darth Vader doesn't have a thing on The Phone Company.

Re:Darth Verizon

[Jim_Maryland]

I'm sure at some point that the rates will change but at least for now, they don't have any text in the plans that say rates will go up after one year of service. Comcast, on the other hand, states that the promotional rates are only good for the promotion period and regular rates apply after that. I'm just glad that we have more competition for these services.

Re:Darth Verizon

[virtual_mps]

I have some experience dealing with AT&T ^W New England Telephone ^W^W^W NYNEX ^W Bell Atlantic ^W^W Verizon. I can say with assurance that there *is* a plan to increase the rates. You just may not be aware of it yet.

Around here Verizon's DSL pricing has decreased over the past few years, while the data rates have increased.

Re:Darth Verizon

[Intron]

The average cost of 1Mbps of residential bandwidth in North America is six times as expensive as Japan or South Korea and four times higher than France

Now please practive saying: Yes, Lord Vader

whoa

[NaklsonofNakkl]

Man, that is insane, i am going to move up to Fiber soon enough but i don't know, i am interested in testing this 100GBps out, i don't doubt it, but i would like to see it myself in action. And maybe to that to my network.

couter strike?

[Cilibrin]

That's what you have to look forward to after marriage. Hee Haw.

AAARGH!!! 100GBps???

[guruevi]

I'll just sit around lurking on Slashdot with 56k. Why can they invent 100GBps but can't even give me a 512k or 1Mbps line?

Re:AAARGH!!! 100GBps???

[Danzigism]

good point dude.. that shit is ridiculous.. i'd like to try getting 100Gpbs internal network speeds.. are there switches capable of such things or would it be more apropriate having direct links from server to server or node to node??

thats what i thought this article was all about at first, then i read the part about it going from Tampa to Houston.. it still amazes me that people are stuck with dialup.. we use it at work, shared with 3 employees haha.. its crazy.. but what can ya do? no cable, no dsl.. and what tight-budget boss would spend money for a satellite setup? bah.. i wish Verizon would of kept their damn promise..

Re:AAARGH!!! 100GBps???

[Anonymous Coward]

Yeah, it sucks doesn't it. One way to at least beat dialup speed, you can look at cellular data.

          CDMA providers will have 1X (144kbps max), EVDO (3mbps? max down, still 1x-style 144kbps up) and EVDO Rev A (5mbps max, and upload speed is 1-2mbps.) Verizon Wireless and Sprint both have 1X everywhere they have service, EVDO in some areas. A fair bit of the EVDO areas are getting/have Rev A now, and then the "old" (3 years at the most..) EVDO equpiment can be reused to upgrade areas that don't have EVDO yet. I have Verizon Wireless' service and realistically get 80-128kbps on 1X, got 1.5-2.5mbps on EVDO when I was in Madison, and got like 800kbps at the worst in New York's EVDO areas.

          GSM providers have GPRS, EDGE, and UMTS. GPRS and EDGE both claim umm.. exagerrated peak speeds assuming you can use a large number of timeslots. If the local GSM network is heavily loaded, you can get fewer timeslots and so slower service than if it's lightly loaded. I've heard GPRS typically might get 40kbps (don't bother, just keep the dialup). EDGE can get 80-250kbps maybe? I'd guess comparable to 1X. I've heard CDMA fanboys claim EDGE is much slower, GSM fanboys claim EDGE is much faster, and others claim they're comparable. I heaven't tried both so I can't say. UMTS can supposedly peak at 30mbps or somthing but existing equipment is in the EVDO to EVDO Rev A speed range. Cingular has mostly EDGE, some GPRS in a few areas (I think? maybe it's all EDGE), and UMTS in a few ares. T-Mobile (probably cheapest data available, if you can get it) last I heard was officially mostly GPRS, but really have EDGE running in many areas they claim are still GPRS.. (weird huh?)

Cool

[funkdancer]

Now all I need is a 100Gb/s hard disk!
Hell, I'd settle for a 1GB/s one!!

Stevens was wrong

[Plutonite]

I think this kind of connection is too much for the tubes. 100 Gbps is more like a truck, I'd wager. A really big truck.

Time your download speed

[kazad]

I've whipped up a calculator to estimate download time:

Download Time Calculator

This is all theoretical of course; networking gurus can add in the effect of TCP headers, etc. to get a real throughput rate. Feedback is very welcome, thought you'd find this useful.

This has nothing to do with TCP...

[Anonymous Coward]

That paper on TCP packet reordering is not related to what was done here.

The 100GE MAC was implemented at the data link layer. We have yet to publish our design for the MAC.