Ethernet's 400-Gigabit Challenge Is a Good Problem To Have

← Back to Stories (view on slashdot.org)

Ethernet's 400-Gigabit Challenge Is a Good Problem To Have

Posted by Unknown on Wednesday October 16, 2013 @02:10AM from the something-about-too-many-cooks dept.

alphadogg writes "As it embarks on what's likely to be a long journey to its next big increase in speed, Ethernet is in some ways a victim of its own success. Years ago, birthing a new generation of Ethernet was relatively straightforward: Enterprises wanted faster LANs, vendors figured out ways to achieve that throughput and hashed out a standard, and IT shops bought the speed boost with their next computers and switches. Now it's more complicated, with carriers, Web 2.0 giants, cloud providers, and enterprises all looking for different speeds and interfaces, some more urgently than others. ... That's what the IEEE 802.3 400Gbps Study Group faces as it tries to write the next chapter in Ethernet's history. ... 'You have a lot of different people coming in to the study group,' said John D'Ambrosia, the group's chair, in an interview at the Ethernet Alliance's Technology Exploration Forum in Santa Clara, California, on Tuesday. That can make it harder to reach consensus, with 75 percent approval required to ratify a standard, he said."

75 comments

Min score:

Reason:

Sort:

Needs more context by Anonymous Coward · 2013-10-16 02:13 · Score: 1, Interesting

So whats the problem? Fitting more bandwidth onto a CAT5 cable? I feel like the summary needs more context.
1. Re:Needs more context by mwvdlee · 2013-10-16 02:22 · Score: 2, Insightful
  
  As I understand it, the problem is more like how to fit 400 1Gbps cables into a single wrapper.
  That, and too many conflicting commercial interests.
  
  --
  Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
2. Re:Needs more context by Anonymous Coward · 2013-10-16 02:37 · Score: 0, Interesting
  
  But we already have 10, 40 and 100gbps cables. Why 400x1?
  Also cat5 is pretty antiquated for the folks who need these speeds. Nobody is really even doing 10gigE over cat5 (incl 6,7,8) for more than tiny patch cables. Fiber is fairly cheap.
3. Re:Needs more context by K.+S.+Kyosuke · 2013-10-16 02:41 · Score: 4, Funny
  
  But we already have 10, 40 and 100gbps cables. Why 400x1?
  Because interface bonding is like herding CATs.
  
  --
  Ezekiel 23:20
4. Re:Needs more context by Anonymous Coward · 2013-10-16 02:45 · Score: 5, Insightful
  
  The problem is that different users have different requirements: Some will want low power requirements but don't need much range. Some will want more range but don't care much about power consumption and cost of advanced signal processing. Up to 1Gbps, Ethernet was a one-size-fits-all standard, mostly because everyone needed roughly the same: cheap, fast and uses existing cabling as much as possible (implying roughly the same range). Technological advances didn't require the kinds of tradeoff that are necessary now. From 10Gbps onward, Ethernet users have become more diverse and the technical challenges have forced more tradeoffs.
5. Re:Needs more context by g0bshiTe · 2013-10-16 02:46 · Score: 1
  
  /rim shot
  
  --
  I am Bennett Haselton! I am Bennett Haselton!
6. Re:Needs more context by Anonymous Coward · 2013-10-16 02:51 · Score: 0
  
  As I understand it, the problem is more like how to fit 400 1Gbps cables into a single wrapper.
  Fiber Optics?
7. Re:Needs more context by luiscolorado · 2013-10-16 03:09 · Score: 2
  
  10, 40, and 100gbps?
  Didn't you mean 10, 40, and 100 Mbps? Or I missed the joke?
8. Re:Needs more context by mlts · 2013-10-16 03:10 · Score: 3, Informative
  
  At the 10gigE point, things diverge. There is single mode ("don't look at laser with remaining eye") media which is great for long distances, but more expensive, multi-mode which is good enough for inside the server room, and good ol' copper. However, this is what SFP modules are for.
  It would be nice if fiber optic made it to the home, other than S/PDIF connections, and preferably with a more idiot-resistant connector than what existing fiber uses, especially with fouling lightpipes due to dust and such. Copper is useful, but eventually for faster connections, we will have to jump ship completely to fiber.
  Of course, once we get 400Gbps, there will be the issues of how it filters down and all the switching/routing fabric needed. Most companies were dragged kicking and screaming to 1Gbps, and might use 10gigE for their trunk, or perhaps their SAN fabric. Trying to get them to 400Gbps for anything other than maybe storage will take a very long time.
9. Re:Needs more context by Anonymous Coward · 2013-10-16 03:21 · Score: 4, Interesting
  
  Nobody is really even doing 10gigE over cat5 (incl 6,7,8) for more than tiny patch cables. Fiber is fairly cheap.
  Your sort of right, but completely wrong. The majority of the 10G ports previously sold are SFP based, but with 10G optical SFPs running ~$80 the vast majority of those ports have direct attach copper cables (CU/CR/CX1/ whatever you want to call it).
  Because of this, the growth direction seems to be towards 10Gbase-T which can do 100 meters over cat6A driving the per port costs down significantly over SFP based solutions. Frankly, with the 10G uptake as a server interconnect, 40G is a good "low cost" switch interconnect/long haul medium. Again killing the need for 10G SFP solutions.
10. Re:Needs more context by SuricouRaven · 2013-10-16 03:22 · Score: 2
  
  What most homes have isn't copper. It's copperish, with traces of iron, tin, water, corrosion, bird droppings and dead rat. Those lines were not made for data. It's a wonder engineers have managed to cram bits down them as fast as they have with DSL.
11. Re:Needs more context by skids · 2013-10-16 03:25 · Score: 1
  
  It would be nice if fiber optic made it to the home, other than S/PDIF connections, and preferably with a more idiot-resistant connector than what existing fiber uses, especially with fouling lightpipes due to dust and such. Copper is useful, but eventually for faster connections, we will have to jump ship completely to fiber.
  I doubt fiber will ever make it in the home market aside from storage attachment. The only way to persuade a typical commodity user to plug anything in these days is if they can charge their battery of it. Will likely see penetration of PoE,PoE+,etc and 10GBase-T, but not much beyond that.
  
  --
  Someone had to do it.
12. Re:Needs more context by Anonymous Coward · 2013-10-16 03:26 · Score: 1
  
  We're going to need a bigger box.
13. Re:Needs more context by raculot · 2013-10-16 03:36 · Score: 0
  
  10, 40, and 100gbps are the ethernet standards generally used in datacenters today. 400gbps would be a reasonable next step for high speed data over copper.
  
  http://en.wikipedia.org/wiki/10_Gigabit_Ethernet
  http://en.wikipedia.org/wiki/100_Gigabit_Ethernet
14. Re:Needs more context by Anonymous Coward · 2013-10-16 04:12 · Score: 1, Insightful
  
  CAT6 and CAT6A are not the same, with 6A its does 100M
  Not sure what you were reading on wikipedia but both
  http://en.wikipedia.org/wiki/Category_6_cable
  and
  http://en.wikipedia.org/wiki/10-gigabit_Ethernet
  as does the IEEE spec.
15. Re:Needs more context by Kjella · 2013-10-16 04:29 · Score: 1
  
  I doubt fiber will ever make it in the home market aside from storage attachment. The only way to persuade a typical commodity user to plug anything in these days is if they can charge their battery of it. Will likely see penetration of PoE,PoE+,etc and 10GBase-T, but not much beyond that.
  I doubt anything has a future in the home, to the home it'll be fiber (23% here in Norway now and rapidly rising) that plugs into a box in the closet that splits it off into TV, phone, wireless and copper wire internet service and so on. GigE over copper is plenty for in-home distribution, even for compressed 4K material unless you've got a big family all watching different things with quad-BluRay quality. Anywhere you're likely to want an Ethernet port you have wall sockets, so no point in powered varieties. Nobody cares about having a 10GbE home network and it'll take forever until you have >1GBit internet connections. Now I'm not going to go 640kB on you and say forever, but for the next decade I see absolutely no demand for anything more.
  If anything the trend is the other way around, you supply power and everything else is wireless even over short distances. That the latest standards have poor range and don't penetrate obstacles well means they work better in apartment buildings due to less interference. Actually in my building it works so well that I'm starting to wonder if it's deliberate, non-interfering materials in apartment-internal walls and blocking materials in walls to adjoining flats. Doesn't seem to have any effect on cell phone signals though, but I receive my own wireless AP extremely well and the others much weaker - I suspect out the window to other buildings in sight. It's the interference that kills wireless performance.
  
  --
  Live today, because you never know what tomorrow brings
16. Re:Needs more context by mjwalshe · 2013-10-16 04:35 · Score: 1
  
  And in some unlucky places Au and Not Cu was used for the last mile due to a spike in copper prices.
17. Re:Needs more context by TheLink · 2013-10-16 04:43 · Score: 3, Insightful
  
  Which unlucky places are those? Insert suitable carrier lost jokes...
  
  Do you mean Al and not Au (Gold)?
  --
  
  Too many replies beneath your current threshold
18. Re:Needs more context by SuricouRaven · 2013-10-16 05:44 · Score: 2
  
  Aluminium isn't a terrible cable - it's got about 60% the conductivity of copper. The problem is joining it. That oxide layer means that any type of twist or post connection is going to make terrible contact. You have to solder it, and it doesn't take solder at all well.
19. Re:Needs more context by AlphaWolf_HK · 2013-10-16 06:01 · Score: 2
  
  Fiber can be cheap, but usually not. We're not just talking about buying actual cables here, we need to be able to buy cable by the spool and terminate it ourselves. Fiber is a bitch to do that (you need gloves, special glue, a polishing pad, and a bunch of other things I can't think of at the moment, plus about 8 minutes of time per termination if you're fast at it, not to mention the possibility of getting glass stuck in your fingers) and doing it on a regular basis eats up your time, and time is money.
  Twisted pair is easy though, I can do it literally in under 45 seconds per end with just two tools: scissors and a crimper.
  
  --
  Careful with names containing L slashdot.org/~AiphaWolf_HK slashdot.org/~AlphaWoif_HK slashdot.org/~AiphaWoif_HK
20. Re:Needs more context by Anonymous Coward · 2013-10-16 06:03 · Score: 0
  
  Actually in my building it works so well that I'm starting to wonder if it's deliberate, non-interfering materials in apartment-internal walls and blocking materials in walls to adjoining flats.
  It probably is, but doesn't have anything to do with wi-fi. Think "firewall", in the traditional sense of a wall that will block (at least for a while) fire. Interior walls usually aren't fireproof, but walls between apartment or condo units generally are (or should be). Okay to burn yourself out, but not your neighbors.
  That the denser (probably concrete, maybe reinforced) walls also block wifi better than drywall does is a bonus.
21. Re:Needs more context by skids · 2013-10-16 06:06 · Score: 1
  
  Anywhere you're likely to want an Ethernet port you have wall sockets, so no point in powered varieties.
  I think here you may be underestimating the level to which the home consumer will finick over convenience. Having one universal power adaptor (an RJ-45 cable) lying around that any gadget can charge off with the added benefit of more reliable low-latency performance beats buying a bunch of wall warts for each device or carrying said wall wart around with you. I could see inductive pads competing, but not wall outlets.
  
  --
  Someone had to do it.
22. Re:Needs more context by mjwalshe · 2013-10-16 06:14 · Score: 1
  
  But its sucks for ADSL deployment
23. Re:Needs more context by TheSync · 2013-10-16 06:17 · Score: 2
  
  I am not seeing much uptake in 10GBASE-T in the data center (yet). People seem to be doing SFP+ copper in a rack (or 10Gbps backplane copper in blade systems), and SFP+ fiber between racks.
24. Re:Needs more context by SuricouRaven · 2013-10-16 06:37 · Score: 1
  
  That's because it isn't soldered. It wasn't installed for ADSL, an twist-connections are good enough for voice, so why spend time doing better? No-one at the time ancitipated it would be use for broadband signals.
25. Re:Needs more context by Anonymous Coward · 2013-10-16 06:52 · Score: 0
  
  Except that RJ45 is already to fat for most devices being carried around. Even on my cheezy little laptop the RJ45 connector is carefully placed in the thickest portion of the body. The connector is going to have to change (along with the plastic clips that break) if Ethernet is to be the universal charging/etc port. That isn't to say, that a 10Gbit PoE network cable isn't a bad idea as a replacement for USB3.
  Running a full blown network stack is actually a good idea, but right now you can buy 5Gbit USB hardware for less than the cost of a SFP or cable on 10Gbit ethernet. Blame the Ethernet vendors for demanding 1000x margins.
26. Re:Needs more context by Anonymous Coward · 2013-10-16 07:41 · Score: 0
  
  Twisted pair is easy though, I can do it literally in under 45 seconds per end with just two tools: scissors and a crimper.
  Not if you plan to keep it compliant with specifications Cat6 or above. It's not trivial to terminate cables with Cat6, 6A, and prohibitively ridiculous at Cat7.
  Cat5 / 5e you could pretty much lick and tie the ends and wrap some tape around it and it would still work. 6, 6A, 7 have nearly exponentially-increasing tolerance requirements to meet specifications.
  I had this discussion 2 years ago with a data center planning manager -- are you **REALLY** saving yourself anything by toiling to terminate cables and keep them compliant when you can just buy prefabricated cables in almost any length you need? That's why they make cable management for racks. If you need a cable that's exactly 21" long, I think it will be OK and the world will not end if you buy a 2ft cable. What you save in time terminating, testing, etc. will more than make up for the "higher cost" of pre-fab cables.
27. Re:Needs more context by drinkypoo · 2013-10-16 12:39 · Score: 1
  
  Aluminium isn't a terrible cable - it's got about 60% the conductivity of copper. The problem is joining it. That oxide layer means that any type of twist or post connection is going to make terrible contact. You have to solder it, and it doesn't take solder at all well.
  The solution in house wiring is to terminate your Al with a small length of Cu. They used aluminum in mobile homes for a while to save weight. I've seen it cause an outlet fire. The solution is to get some copper ends which get attached to the aluminum wiring and then sealed on. I've never actually seen the fix done, I think they're crimped on and then either protected with a compound or with an epoxy or glue. I wonder if the same couldn't be done for existing data wiring.
  
  --
  "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
28. Re:Needs more context by guruevi · 2013-10-16 14:34 · Score: 2
  
  You can count the contractors that do proper CAT6A wiring on one hand. You can barely get a contractor to do CAT5E correctly (without using wire nuts to keep 2 ends together), let alone CAT6/CAT6A/CAT7, heck I have a hard enough time buying decent quality copper CAT5. CAT6 cable comes in at ~$300/1000ft, CAT7 at ~$800/1000ft. Decent quality MM fiber comes in at ~$200/1000ft.
  
  --
  Custom electronics and digital signage for your business: www.evcircuits.com
29. Re:Needs more context by Anonymous Coward · 2013-10-16 17:09 · Score: 0
  
  If I read the summary correctly (not the article FTW), the problem is getting 75% of the new, larger group to agree on a single standard.
400 GB is easy by davidwr · 2013-10-16 02:17 · Score: 2

400 GB per small* unit of time on the other hand ...
* "small unit of time" being 1 second or less

--
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
1. Re:400 GB is easy by somersault · 2013-10-16 02:24 · Score: 1
  
  Gb*
  *sorry
  
  --
  which is totally what she said
2. Re:400 GB is easy by Anonymous Coward · 2013-10-16 03:00 · Score: 0
  
  On the other hand what?
3. Re:400 GB is easy by Anonymous Coward · 2013-10-16 04:23 · Score: 0
  
  He kind of put the first sentence in the topic so what follows is the opposite of it, which is "is hard".
Use the old standard by Anonymous Coward · 2013-10-16 02:18 · Score: 0, Troll

How hard can this be ?
Use the old standard, just make it faster.
Problem solved.
1. Re:Use the old standard by jones_supa · 2013-10-16 04:31 · Score: 1
  
  Exactly. You just take the mainboard of a 1Gbit router, locate all the clock sources and replace them with 400 times as fast oscillator. Then you turn the power back on, crack open a beer, and enjoy your new intertubez with a wide grin.
Patents by Dachannien · 2013-10-16 02:34 · Score: 5, Insightful

The problem isn't that you have a bunch of squabbling engineers who can't even figure out how to split a lunch check. It's that you have a bunch of executives and attorneys that want to get as much of their company's IP piled into the standard as possible.
1. Re:Patents by Anonymous Coward · 2013-10-16 03:36 · Score: 0
  
  That too. But someone not from the field would be surprised about the amount of choices and tradeoffs involved. And that's highly technical and at the same time extremely critical (always to just a subset of the potential users :-( ) matters, such as power costs, silicon area/complexity costs, latency, robustness (signal integrity)...
  It was quite easy for optical 1000BASE-X, as ASK modulation is trivial. It was still damn easy for 10GBASE-X optical (can still use very simple line modulation) and it is still an ongoing issue on the electrical side (SFP+ direct-attach / 10GBASE-CX4 / 10GBASE-T), where you ended up with two successes (SFP+ for any latency or power-critical application and any cost-critical application that can live with SFP+-DA cable-length requirements, 10GBASE-T for everything else).
  100GbE has no Cu-based soltion worth mentioning, and it was nasty on the line modulation side even for optical, to the point that people are still looking for better solutions than the current ones as they are too expensive *and* don't cather for all markets. And it ended up being *slow* (high latency) for some critical applications. It gets worse when you need to kick up the FEC (which you most often do outside in-datacenter distances).
2. Re:Patents by Anonymous Coward · 2013-10-16 03:58 · Score: 0
  
  This.
  This is the reason we still use DNS, a protocol with no real security and no real authentication.
  This is the reason we still use SMTP, a protocol with no real security and no real authentication.
  We're stuck with these standards because there are a bunch of IP laywers that will parasitically destroy the economic viability of any new standard with vaguely worded patents.
3. Re:Patents by bob_super · 2013-10-16 04:02 · Score: 1
  
  No, the main problem is that 40Gb/s laser modulation is expensive, anything above is not out of the labs yet.
  And the 25Gb/s electrical modulation which is supported by FPGAs driving those lasers will run you into the 5-figures BOM cost in a blink.
  At 400Gb/s today, you need 16 x 25G links (bi-di, that's 64 critical traces), which has to be done very carefully and also dictates powerful redundant cooling.
  Internally, a 400Gb/s bus at a rate supported by general silicon is pretty wide (512b in ASIC, 1024 in FPGA) and requires a lot more supporting logic than 4 times 100G did.
  Really, the only problem is that we have to wait a bit for Moore to catch up every time we multiply by 4.
4. Re:Patents by Anonymous Coward · 2013-10-16 04:37 · Score: 0
  
  This.
  This is the reason we still use DNS, a protocol with no real security and no real authentication.
  This is the reason we still use SMTP, a protocol with no real security and no real authentication.
  We're stuck with these standards because there are a bunch of IP laywers that will parasitically destroy the economic viability of any new standard with vaguely worded patents.
  Which lawyer has gone after Namecoin? Its issues are purely internal* economics, not legal or technical.
  * By this I mean its own system of auctioning off all the big names several years ago before anyone heard of it, and then charging less than a penny for domain squatters to renew. IBM probably won't support an open standard where they have to pay out the ass to own IBM.bit.
5. Re:Patents by Shatrat · 2013-10-16 04:59 · Score: 1
  
  We've got 100Gbps optics live on our network. Granted, even the Coherent stuff is broken down into two polarized carriers or two different wavelength carriers that are modulated at 50Gbps using QPSK. It's expensive, but it's cheaper than the equivalent number of 10Gbps circuits.
  
  --
  09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
The faster data moves by Anonymous Coward · 2013-10-16 02:45 · Score: 0

The more I'm going to have to pay for the privilege.
1. Re:The faster data moves by Tynin · 2013-10-16 03:32 · Score: 1
  
  The more I'm going to have to pay for the privilege.
  Pretty much false. I remember in the 90s a T1 costing ~$700 for the ISP, and ~$700 for the telco fees, and only giving 1.5 Mbps. Nowadays it is common for households to have 10x that speed down stream and roughly the same speed for upstream, for ~1/20th the cost.
2. Re:The faster data moves by iggymanz · 2013-10-16 03:49 · Score: 2
  
  not a valid comparison, since a business grade symertric medium of course has a premium A T1 compariable line still is $300 month, and plenty of businesses have them for their mission critical traffic. Company I last worked at had 40 Mbit Comcast business line with routed subet, but they still had the T1 backup for email and main web site and replication to offsite DNS. That Comcast line went down or flapped sometimes, but the T1 was always there
3. Re:The faster data moves by Pieroxy · 2013-10-16 03:59 · Score: 1
  
  My DSL line happily uploads 1.5Mbps for under $30 per month...
  
  --
  Write boring code, not shiny code!
4. Re:The faster data moves by Anonymous Coward · 2013-10-16 04:17 · Score: 0
  
  You're doing it wrong. T1 lines are not that reliable. Cable + DSL is a good alternative. You get speed benefits without the cost or complexity of a T1.
  A few years back, I migrated my employer off a T1 line to cable business internet through comcast with a DSL backup. Our T1 line dropped every few days and it was a blessing to go to DSL.
5. Re:The faster data moves by Ravaldy · 2013-10-16 04:45 · Score: 1
  
  Yeah but you get E10 (10Mbps up/down) for less than $800/month. That's 10 times the speed for the same price at T1s in the 90s. T1s are still expensive because of their large overhead and expensive equipment.
6. Re:The faster data moves by isorox · 2013-10-16 04:55 · Score: 1
  
  not a valid comparison, since a business grade symertric medium of course has a premium A T1 compariable line still is $300 month, and plenty of businesses have them for their mission critical traffic. Company I last worked at had 40 Mbit Comcast business line with routed subet, but they still had the T1 backup for email and main web site and replication to offsite DNS. That Comcast line went down or flapped sometimes, but the T1 was always there
  Most businesses I've seen are moving to MPLS. We have a T3 we're trying to migrate away from as we can replace it with 2 50mbit mplss and make a 70% saving, and replace the kit with something that is actually supported.
  One problem we're having is we have 2 E1 lines on the barer, but you can get E1 over IP converters which look promising.
7. Re:The faster data moves by Shatrat · 2013-10-16 05:11 · Score: 1
  
  Until the oversubscribed DSLAM starts dropping packets at 3PM. Jitter and latency will be worse as well. T1 is dedicated bandwidth to the router or switch.
  
  --
  09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
8. Re:The faster data moves by Anonymous Coward · 2013-10-16 05:40 · Score: 0
  
  > Most businesses I've seen are moving to MPLS.
  Awesome! I live in Minneapolis and I'm looking for a new job...
9. Re:The faster data moves by iggymanz · 2013-10-16 07:21 · Score: 1
  
  consumer grade, no service guarentees and not 100% reliable, that's the big difference.
  I see other similar comments, it's like when people see the specs on the computer controlling a space rover and they say "but my desktop is so much faster!"
10. Re:The faster data moves by iggymanz · 2013-10-16 07:23 · Score: 1
  
  that's not normal, cable and DSL are very shaky compared to typical T1
11. Re:The faster data moves by iggymanz · 2013-10-16 07:33 · Score: 1
  
  E10? in the UK for ITU-T they have E1 through E4.....we're talking about business grade time division multiplex carrier lines, not DSL or cable or other consumer grade shakier and less reliable tech
12. Re:The faster data moves by iggymanz · 2013-10-16 08:18 · Score: 1
  
  MPLS works over a number of carrier technologies including T1, E1, ATM, Frame Relay, and DSL.
13. Re:The faster data moves by Anonymous Coward · 2013-10-16 08:39 · Score: 0
  
  because you never can over subscribe a router......
14. Re:The faster data moves by Methlin · 2013-10-16 08:47 · Score: 1
  
  Shhh, don't tell anyone, but a T1 is just two (or one, depending on span type) DSL circuits and has all the same impairments and limitations as the first two-wire residential DSL standard; which based itself off of two-wire T1 spans. The difference of course is repeaters for T1s are common place every 3000-5000ft. They exist for xDSL as well but are far more expensive and only ever seen in rural areas.
15. Re:The faster data moves by isorox · 2013-10-16 09:17 · Score: 1
  
  MPLS works over a number of carrier technologies including T1, E1, ATM, Frame Relay, and DSL.
  They do (and other technologies. I wonder if you can do MPLS over MPLS. I have 3 days put aside next month for some fiddling with MPLS and VPLS), but as all I get is an ethernet handoff on a piece of fibre, it means that old inefficient technologies like T1/DS3/etc can be bypassed.
  From my point of view, I pay $xxx per month, and get a cable in my equipment room, the fibre heads off to somewhere, the ethernet packets are tagged with MPLS and shifted quickly and reliably across multiple high bandwidth links (Say SEA ME WE 4) before landing in my equipment room 6,000 miles away unmolested. I'm not sure what 1tbit intercontinental fibre runs over, I'm sure its very complex and expensive, but that cost is shared amongst 10,000 other customers.
  From my point of view as an end user though, I no longer need half a bay of ancient equipment to deliver a 45mbit bearer, I have a fibre landing on my £300 router at each end, I run an IP circuit on a /30 between the two routers, and I drop vlans off each side, and shape as much as I want. The capital cost drops 2 orders of magnitude, the reliability jumps up, the support costs collapse, and the flexibility increases.
  Why would I, as a non network service provider, ever think of anything other than an ethernet or IP line?
16. Re:The faster data moves by WuphonsReach · 2013-10-17 08:56 · Score: 1
  
  Eh, our T1 goes out maybe once per year, and has a SLA.
  
  Our cable network drops 2-3x per month and we have no better options. But it's about 30-35Mbps inbound and 3-5Mbps outbound, so far better bandwidth then the T1.
  
  So we use both. The T1 is the fallback line for the cable internet and traffic automatically shifts from the cable line to the T1. It might take 2-3 minutes for the shift to happen, but its automatic and is better then a 30-240 minute downtime.
  
  --
  Wolde you bothe eate your cake, and have your cake?
17. Re:The faster data moves by Cyberdyne · 2013-10-18 23:20 · Score: 1
  
  E10? in the UK for ITU-T they have E1 through E4.....we're talking about business grade time division multiplex carrier lines, not DSL or cable or other consumer grade shakier and less reliable tech
  I imagine 'E10' there is a reference to 10 Mbps metro Ethernet, something like the Ethernet in the First Mile approach. There's nothing inherently "consumer grade" about DSL itself: indeed, even E1 "leased lines" get delivered over HDSL or similar in some cases. Unlike cable, which is contended and prone to collisions, DSL gives you a constant bitrate (unless configured to vary to squeeze higher bitrates when line quality permits) point to point link, just like a conventional leased line - all the performance fluctuations of typical DSL Internet access come further into the network, where your 20 Mbps connection is sharing a 1 Gbps backhaul with a thousand others and gets choked up when everyone is streaming X-Brother Get Me Out Of Here or whatever. Give the DSL link dedicated or uncontended backhaul like leased lines have, you'll get the same performance too.
Ethernet should concede to Infiniband by Anonymous Coward · 2013-10-16 03:55 · Score: 0

For short ranges, Infiniband is winning the high speed data transfer race. Infiniband is designed for ~10 meter, low latency, >10s gb/s data. Infiniband is designed for performance, Ethernet was not. Yes, Inifiniband is more complex than Ethernet, but at the gain of performance. Ethernet should recognize that ease of use, and ubiquity, and not performance, is its strength, and Ethernet should not compete with Infiniband in the very high end.
1. Re:Ethernet should concede to Infiniband by Anonymous Coward · 2013-10-16 05:03 · Score: 0
  
  Its also because the hardware is cheap compared with cisco/juniper/etc ethernet. The big name ethernet vendors are fucking themselves.
  Now that commodity switching chips are out, you can get netgear 10gbit switches for literally 1/10th to 1/50th the cost of similar cisco garbage.
  Its like, why can I buy a tablet for $300 that has a better screen than the $1000 ultrabook from $PCMANUFACTUER?
2. Re:Ethernet should concede to Infiniband by MachineShedFred · 2013-10-16 07:31 · Score: 1
  
  Yes, you can get a cheap 10GbE switch. No, it doesn't have nearly the management that a Cisco or HP switch has.
  If you are using one or two, you probably don't care. If you are using one or two thousand of them, you VERY much care.
  
  --
  Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
3. Re:Ethernet should concede to Infiniband by Anonymous Coward · 2013-10-16 09:36 · Score: 0
  
  I'm curious what feature you think is available on the ciscos that everyone needs that isn't available in the netgear layer3 switches, or what features you really need in a top of rack L2 switch?
  AFAIK, non of the really big data centers are using cisco/etc gear for TOR. Heck it sounds like a number of them have custom switches they designed themselves.
fibre types by Anonymous Coward · 2013-10-16 04:31 · Score: 0

I'm curious to know if 400+ Gb/s will be able to use existing OS1/2 and OM3/OM4 fibres, and at what distances.
the cool slang words like 4GOOB by Anonymous Coward · 2013-10-16 04:50 · Score: 0

g00ber-fi ultraG00B-HD
droidGUbE
iG00b.
4GOOB-LtE (which will be proprietary), 4GOOB-max (really runs at 10g) 4GOOB-J1000G *might* give you original spec speed if nobody in your cell is on
Microsoft by StormReaver · 2013-10-16 04:59 · Score: 1

Before you let anyone into your standards committee, make sure they don't work for Microsoft or a Microsoft affiliate. And if they do, make sure Microsoft isn't trying to push through a competing "standard".
Be sure you learn the ISO's lessons regarding Microsoft and its henchmen stuffing standards bodies.
1. Re:Microsoft by disposable60 · 2013-10-16 05:19 · Score: 1
  
  Sorta' like Micron did jamming RAMBUS-patented tech into the DDR-something definition?
  
  --
  You're looking for quotes? See my journal.
Bottleneck by Anonymous Coward · 2013-10-16 14:48 · Score: 0

I think most people are missing the point. Everyone wants more speed. Most people agree that speed limits on highways should be increased. Same with networking. However, nobody wants the rural roads or suburbs to have faster speed limits than the highway. The highways would become bottle necks more than they already are.
You can't have servers attached at 400G when your switching fabric needs to handle ~n X (number of ports). What is the point of going faster for small business if their switch just drops packets due to a slower switching fabric AND that's just internal. Start adding WAN's and you need not only faster switches but bigger buffers.... and nobody likes buffer bloat! This might work well for a backbone infrastructure (switch to switch, etc) but I'd never (personally) put servers on this until the back-end could support it.