Slashdot Mirror

← Back to Stories (view on slashdot.org)

IEEE Sets Sights on 100G Ethernet

Posted by Zonk on from the aint-no-mountain-high-enough dept.

coondoggie writes to mention a Network World article about the IEEE's new 100G Ethernet initiative. The organizing body's High Speed Study Group has voted to try for the 100G standard over other ideas, like 40Gbps ethernet. From the article: "The IEEE will work to standardize 100G Ethernet over distances as far as 6 miles over single-mode fiber optic cabling and 328 feet over multimode fiber. With the approval to move to 100G Ethernet, the next step is to form a 100G Ethernet Task Force to study how to achieve a standard that is technically feasible and economically viable, says John D'Ambrosia, chair of the IEEE HSSG, and scientist of components technology at Force10 Networks." With video download services and interactive media becoming ever more the focus of internet startups, the organization is eager to offer a way to aggregate pipes in the coming years. The current thinking is that achieving these speeds will be reached by advancing bonding techniques for 10G signals over multiple fibers.

111 of 136 comments (clear)

  1. I'm going to guess... by Anonymous Coward · · Score: 1, Insightful

    That off the shelf hardware won't be able to saturate a 100Gb connection.

    1. Re:I'm going to guess... by Ajehals · · Score: 3, Insightful

      Chicken and egg -

      Once the connectivity is there, hardware will become available and gradually more accessible as it is taken up, same goes the other way - if someone suddenly comes up with a bus and card capable of even higher speeds, it will slowly become available and more accessible until connectivity catches up and everyone wants it. Its all about getting to the point were a (potential) mass market appears and it makes the R&D viable. In the short term you will obviously see niche markets for it anyway - and they will pay buckets of cash for this kind of tech because they see a benefit from it.

    2. Re:I'm going to guess... by jeffmeden · · Score: 5, Funny

      Hi I'm Progress, and I'm going to guess that we haven't met. I will be forever pushing forward with faster speeds. Thought you were happy with Gigabit over copper? LUDDITE! 10Gbit is enough for all your communication needs since you can xfer the library of congress 5 times a minute? THINK AGAIN! 100G Ethernet is the natural progression and before long you WILL want it. Trust me, I have been working this way for thousands of years. Glad we could get acquainted, now excuse me I need to get back to hiding from politicians.

    3. Re:I'm going to guess... by Shakrai · · Score: 3, Funny

      Glad we could get acquainted, now excuse me I need to get back to hiding from politicians.

      Too late, we found you.

      --
      I want peace on earth and goodwill toward man.
      We are the United States Government! We don't do that sort of thing.
    4. Re:I'm going to guess... by CmSpuD · · Score: 1

      I misread that as "Hi, I'm Congress!" and did a double take. Congress? Progress? That can't be right..

    5. Re:I'm going to guess... by Stachybotris · · Score: 1

      Yes, and as little as 12 years ago I thought that we'd never be able to fill a single CD-Rom.

    6. Re:I'm going to guess... by forkazoo · · Score: 2, Insightful
      That off the shelf hardware won't be able to saturate a 100Gb connection.


      Depends on which shelf.

      Seriously, a lot of folks commenting on this news item seem to be convinved that all networks have only one node. Sorry, but I'm on a university, and I think that our interbuilding connections could really saturate a 10 Gb connection in the near future. It may be a long time before one PC can make use of a 100 Gb connection, but it won't be long at all before 1000 PC's can. Deployments will start the same way that 100 Mb and 1 Gb started. Backbone switches will move to the fastest speed, feeding workstations moving at slower speeds. Some specialised equipment will be available for systems which really need to actually move that kind of data to a particular node.
    7. Re:I'm going to guess... by itlurksbeneath · · Score: 1

      If you're on a 100G network by yourself, then correct. If there are a thousand other computers going through the same backbone, then maybe not.

      --
      Have you ever considered piracy? You'd make a wonderful Dread Pirate Roberts.
    8. Re:I'm going to guess... by nokiator · · Score: 1
      This standard effort appears to be targeted at 2009-2010. By that time, we are likely to see CPUs with 8-16 cores. This implies a total of 16-32 cores on a 1U dual-processor server, 32-64 cores on a 2U dual-processor server. Most server/chipset vendors are adding enhancements to further improve network throughput like TCP offload engines or virtual-NICs. By 2010, it is conceivable that one of these servers could easily sustain multi-gigabit network throughput and burst close to 10Gb/s for short periods of time. Once you connect a set of such servers to an access switch with 10G ports, how do you uplink this much network throughput? What kind of network interfaces and how much switching bandwidth would you need in the switch at the next level of hierarchy?

      By 2009-2010, even if fiber-to-the home is not widely deployed, Cable and DSL customers will most likely have sufficient access bandwidth to support on-demand HDTV downloads. Even if we assume the video is H.264 encoded, you are still looking at 10Mb/s sustained bandwidth per subscriber. It takes only 10,000 subscribers to fill a 100Gb Ethernet pipe at those kind of access rates.

      IMHO, there will be plenty of applications for 100Gb Ethernet.

    9. Re:I'm going to guess... by Gilmoure · · Score: 1

      I remember seeing CD's back in '93. Already, they were too small for my porn collection. Ah, giffy girls; you're so low-rez now.

      --
      I drank what? -- Socrates
    10. Re:I'm going to guess... by WuphonsReach · · Score: 1

      Mmm... 8 cores in 2009? Maybe by 2010. I figure 8/16 core chips will require the move to 45nm. (Unless we're talking non-x86.) We are barely getting into quad-core in 2006 and it won't really be available until 2007. And RAM speeds are still in the 3-4GB/s range (what's next? DDR3? DDR4?).

      (I'm not as optimistic as you are...)

      --
      Wolde you bothe eate your cake, and have your cake?
    11. Re:I'm going to guess... by nokiator · · Score: 1
      We already have quad-core chips in 65nm process. As area scales much better than speed with process technology transition, 8-cores in 45nm would be relatively straight forward in terms of implementation. Note that Intel is already planning to ship 45nm processors in the second half of 2007 and AMD is planning to start the 45nm transition in 2008. I would actually be very surprised if we don't see 8 core processors by 2009.

      The real question is whether applications will be able to take advantage of 8 or more cores. In the server world, this is obviously possible as 8-way/16-way/32-way SMP servers have been around for some time. There are also other multi-core architectures (e.g. Sun) which has proven that servers can take advantage of large numbers of cores. Multiple cores on a standard x86 processor simply enables similar technology at a lower cost point. For desktops, games and video processing (especially encoding) are probably the most CPU-hungry applications. Both class of applications can be parallelized to take advantage of 8 or more cores.

      As for memory bandwidth, it will indeed become an issue with more cores on a single die. However, unlike memory latency, memory bandwidth is a manageable problem. DDR3 chips with 666MHz clock rate (1.33Gb/s data rate per pin) are already sampling. Performance oriented Intel desktop chipsets will transition to DDR3 by the second half of next year. By that time, we are going to 800MHz clock rate (1.6Gb/s data rate per pin) will also be available. This effectively enables desktop chips to double the memory bandwidth for the same number of pins. As for servers, FB-DIMM type of architecture enables memory bandwidth to scale to much larger numbers, albeit at additional cost. The other potential bottleneck in the memory datapath is the system bus, FSB. It is very likely that Intel will get rid of the FSB by 2009 and switch to CPUs with integrated memory controller - similar to what AMD is doing today. Alternatively, it is possible to scale the FSB by using a differential point-to-point architecture. For example, 64-lane PCI-E 2.0 compatible FSB would give you 320Gb/s raw bandwidth in each direction.

  2. What's in it for desktop users? by morgan_greywolf · · Score: 4, Insightful

    As it is, your average desktop will not handle anything even close to 100G Ethernet. At that point, your bottleneck is the PCI or PCI-X bus. As the bus has been one of the slowest PC components to innovate, I see these new, ultra-high speed Ethernet standards as only benefiting backbone providers, etc., for many years to come.

    --
    My blog
    1. Re:What's in it for desktop users? by corsec67 · · Score: 4, Insightful

      What, you think that any ISP would actually allow downloads fast enough to use over 100baseT?

      Really, even full 10baseT (as an obtainable download speed, not just the home->CO link speed) would be an improvement to many people.

      --
      If I have nothing to hide, don't search me
    2. Re:What's in it for desktop users? by UnknowingFool · · Score: 4, Funny

      Well that swimsuit supermodel in the magazine I've been lusting after will never date me, but I can dream about it can't I? At least 100G Ethernet to the desktop might be realized in my lifetime. A supermodel, not so much. :P

      --
      Well, there's spam egg sausage and spam, that's not got much spam in it.
    3. Re:What's in it for desktop users? by pacman+on+prozac · · Score: 1

      They would allow it if it was cost effective, some countries already have 100mbit to the home. To get that requires a huge backbone to start with and it needs to be available to the Telcos/ISPs at reasonable prices.

      Bonding huge links together will be quite a feat, as far as I know the main bonding protocols in use now (etherchannel, LACP, etc) are based on current ethernet standards so may need some reworking, unless the large links are already using Ethernet (DWDM maybe?). Then there's the small matter of getting some hardware together that can switch at 100G....

      But the OP is quite right, this isn't really aimed at end users and they'll only get benefits indirectly. This is aimed at things like British Telecoms new MPLS network that is supposed to carry all voice and data traffic on a single IP network.

    4. Re:What's in it for desktop users? by FireFury03 · · Score: 4, Insightful

      What, you think that any ISP would actually allow downloads fast enough to use over 100baseT?

      Believe it or not, some people use LANs for things other than accessing the internet... The internet connection speed becomes unimportant if the network is actually a SAN.

      Really, even full 10baseT (as an obtainable download speed, not just the home->CO link speed) would be an improvement to many people.

      We're reaching the point now where I've stopped caring so much about download speed (I have an 8Mbps DSL) - upload speed is becoming a serious headache since on most ADSL lines (at least in the UK) it tops out at ~340Kbps. At that upload speed you're talking about ~45ms per MTU sized (1500 byte) packet - that's quite a lot of latency jitter and can cause serious problems for realtime applications such as VoIP, which often have jitter buffers of only around 100ms long.

      --
      http://blog.nexusuk.org
    5. Re:What's in it for desktop users? by Fweeky · · Score: 1

      Well, yes, that's what's in it for desktop users; networks upstream can scale further.

      And who says you have to connect it to PCI-(E|X)? Hook it up directly to a HyperTransport link and talk to other systems on the network at reasonable speeds.

    6. Re:What's in it for desktop users? by GigsVT · · Score: 3, Interesting

      Don't worry about content. We'll find plenty of things to fill that with.

      Look at a platform like Second Life. It uses a very simplified version of CSG 3d modelling because of bandwidth constraints of current broadband. Now imagine Second Life with fully deformable meshes and high resolution textures in a world that is downloaded faster than you can move through it.

      Anyway, we'll find plenty of things to do with more speed, we always do.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    7. Re:What's in it for desktop users? by MojoRilla · · Score: 1

      Sure, your average desktop will not handle 100G Ethernet, but what kind of content or traffic could possibly require that much desktop bandwidth? And since this is over fiber, there are very few desktop networks out there you could even plug the PC into.

      This will certainly be a backbone technology, and a server technology. But this particual technology doesn't seem likely as a desktop technology in the near future.

    8. Re:What's in it for desktop users? by jellomizer · · Score: 2, Informative

      Well a faster Pipe for the ISP allow them to increase max speed for their users. So say they had a 100GB Eathernet for their backbone internet connection. That means they could possible increase the max speed for their customers to closer to 10MB as of right now most U.S. ISP tend to cap to 5MBS Anything above that could be to much demmand on their systems. So you as a desktop can see an impovement.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    9. Re:What's in it for desktop users? by Znork · · Score: 4, Interesting

      Heck, with 100Gb ethernet, who says you have to _have_ PCIe; once you reach those speeds it would be entirely viable to move most PC components to their own ethernet bus/network. Imagine having your NVidia graphics units connected to your LAN and usable from any of your PC's; plugging another unit into the network makes it instantly accessible by all devices as tho it was more or less local hardware. Etc.

      SAN is storage moving that way, we might very well expect other components to move in the same direction.

      Of course, expect a horde of crap patent applications for shit like 'graphics acceleration _over a network_' just because the technology becomes feasible. Which may drive prices through the roof and/or hold development back a decade or five.

    10. Re:What's in it for desktop users? by casualsax3 · · Score: 1

      Right now actually the PCI bus can't even take advantage of a 10G Ethernet card - you'll see that in real world conditions standard server class hardware is pretty much capped at 2G because of bus limitations.

    11. Re:What's in it for desktop users? by monopole · · Score: 1

      But w/ a 100Gb Ethernet you can download the Real Doll data of the supermodel to your desktop fabber real fast!

    12. Re:What's in it for desktop users? by smoker2 · · Score: 1
      - upload speed is becoming a serious headache since on most ADSL lines (at least in the UK) it tops out at ~340Kbps
      Well, apart from the fact that I seem to get 448kbps up (and I'm not alone) you can always try paying a little more.

      For example - these people are offering "Up to 24,576K download speed", and "Up to 1,331K upload speed" on a residential use basis for only £85.47 per quarter (£28.49 per month) I don't know the VAT status of that quote. Or they also offer this.

    13. Re:What's in it for desktop users? by Ryan+Amos · · Score: 1

      Because it's not about the desktop, but yes, desktop users benefit.

      As backbone providers get more capacity, they can deliver faster speeds to their customers (your ISP and the ISPs of the websites you visit.) Ethernet is a very cost-effective physical transport layer as it removes some of the administration headaches involved with point-to-point links. This will eventually drive down the cost of fast backbones, allowing more bandwidth for less money.

      Your average desktop user doesn't have a need for 10G, muchless 100G ethernet. By the time he needs 100G ethernet, there will be a bus capable of supplying it. Right now his hard drive can't even read anywhere close to fast enough to saturate a 1G link, so we have a ways to go in other areas first.

    14. Re:What's in it for desktop users? by russ1337 · · Score: 1

      >>>"As it is, your average desktop will not handle anything even close to 100G Ethernet"

      Key words there "As it is"; If they build it, etc...

      Plus, I'd quite happily have 100G to the house. It would not be for one computer, but for the four that I currently have, plus who knows how many I'd have by the time they roll it out.

      A couple of apps I might use it for [pipedream]:
      thin client gaming to Google-games(TM), where the googleserver does all the game crunching, HDR etc. I might want to Slingbox HD media to my... er,,.. somewhere.
      and, the most usefull: I might need to download pr0n really fast.

    15. Re:What's in it for desktop users? by pikine · · Score: 1

      At some point in your lifetime, 100G Ethernet will be available on the market, and you will be willing to pay for it. Would you ever be willing to pay for a supermodel, which has been on the market since known human history?

      See, that's the problem.

      --
      I once had a signature.
    16. Re:What's in it for desktop users? by Frumious+Wombat · · Score: 1

      And some of us would be happy if they only marginally bumped up the capacity, but worked on the latency issues instead. Something with Gigabit to 10Gigabit speed, but with less than 10 usec of latency would be a good start. Myrinet for everyone, since some of us use Beowulf-type clusters for work, rather than humor.

      --
      the more accurate the calculations became, the more the concepts tended to vanish into thin air. R. S. Mulliken
    17. Re:What's in it for desktop users? by KingMotley · · Score: 1

      I guess I'm not one of them. I can burst up to 35Mb/sec, and can sustain downloads of just over 10Mb/sec (It's not quite 11Mb/sec). Even full duplex 10base-T wouldn't keep up with that.

    18. Re:What's in it for desktop users? by jabuzz · · Score: 1

      News flash, your average desktop cannot even handle a 1Gbps link, let alone a 10Gbps. Experience tells me that you will see about 30MBps max out of a 1Gbps link on desktop hardware. You need server grade kit to go faster. I can max out a dual bonded 1Gbps link on my servers for example.

    19. Re:What's in it for desktop users? by NeuralSpike · · Score: 5, Insightful

      Dude, I think with the various packet headers etc., 100Gbit isn't all that much faster than a 16x PCI express slot. And then there is latency...

    20. Re:What's in it for desktop users? by bill_kress · · Score: 1

      I imagined it--and it's still totally meh.

      Perhaps they could add something more fun than "Buying Stuff" instead. Even Pokemon had a game to play while you were mindlessly collecting "them all".

    21. Re:What's in it for desktop users? by TheSHAD0W · · Score: 1

      Even if the desktop user never has direct access to it, he will benefit. This sounds like a great way to solve the last mile problem; a 100-gigabit line will support 500-1000 users at 100 megabits each.

    22. Re:What's in it for desktop users? by Com2Kid · · Score: 1

      This is quite true, my laptop pegs out its CPU after around 8MB/s. (Pentium M 1.6ghz)

      Not to mention the poor HD, that is not contiguous writing, but rather multiple streams, so I imagine that the poor disk head is jumping all over the place trying to place that data!

      --
      Need help treating your acne? Come here!
    23. Re:What's in it for desktop users? by 19thNervousBreakdown · · Score: 2, Informative

      At 100Gbps, your processor's L1 cache is a bottleneck.

      --
      <xml><I><am><so><damn>Web 2.0</damn></so></am></I></xml>
    24. Re:What's in it for desktop users? by GigsVT · · Score: 1

      You don't "play" SL. That's like saying you got on this "world wide web" but the game of "buying stuff" was too boring. It's a platform, not a game.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    25. Re:What's in it for desktop users? by anticypher · · Score: 1

      ADSL2+ speeds long ago left 10BaseT speeds behind. Now that ADSL2+ is becoming obsolete, and fiber is going in, expect that 3rd world countries will pick up the kit for cheap. 20Mbps sustained downloads are pretty common in Europe, 8-10Mbps in the UK. I know of many companies that have a single ADSL line for the 20 to 50 PCs in the office, its enough bandwidth for casual internet use.

      Of course, if you live in a country with a corrupt administration and a broken telecoms regulator, then you will never know the joy of gigabit speeds or even relatively sluggish 24Mbps.

      the AC

      --
      Hemos is like...sci-fi fans;he thinks technology is cool, but he hasn't bothered to understand the science it's based on
    26. Re:What's in it for desktop users? by Vancorps · · Score: 1

      Wow, what the hell equipment are you running? My laptop is 2.4ghz Pentium M and I can transfer a gig in a minute. That's ~17megabytes/sec. Sounds to me like your network is either congested or your file server is sorely lacking. Hell, with my 14drive SATA array I break 550megabytes/sec of effective throughput on my file-servers. My SAN boasts even more performance so I see this stuff as becoming very useful. I can't utilize 100gbit but 10gbit I could certainly saturate pretty easily. Think VMWare images pushed to the desktop to provide users with a consistent clean work environment.

      Another use for me at least would be transport of raw video so other machines could do the encoding so then I don't have to worry about having hardware do the whole thing in real-time on one box. Fortunately dual Opterons with dual cores provides enough for me to encode 4 streams in real-time but eventually I'll probably want more than 4 streams and a faster network connection would certainly help with that.

    27. Re:What's in it for desktop users? by morgan_greywolf · · Score: 1

      Yep. Mod parent up. Amazing the number of people on Slashdot who have no clue about networking.

      --
      My blog
    28. Re:What's in it for desktop users? by WuphonsReach · · Score: 1

      Pretty sure the PCIe bus can supply 1GByte/s to a 10Gbit card. I know low-end DDR2 RAM is capable of around 3GB/s of sustained data transfers.

      And... I forget what the speeds are for HT...

      --
      Wolde you bothe eate your cake, and have your cake?
    29. Re:What's in it for desktop users? by Com2Kid · · Score: 1

      Wow, what the hell equipment are you running?


      Dell D600 w/ integrated network card.

      4 to 5 transfers at a time.

      --
      Need help treating your acne? Come here!
    30. Re:What's in it for desktop users? by altek · · Score: 1

      Ummmm... typical of a slashdot comment, all that exists is the backbone providers and home users, eh? What about enormous LANs in high-bandwidth settings? Hospitals, publishing companies, graphics design companies, audio engineering companies, research facilities, these are just a VERY few examples of places that would benefit from this.

      And with 6-mile over single-mode fiber, even places with multiple physical sites can benefit. Warehouse is 9 miles away you say? Well, just stick in a device to condition and amplify the signal, and send it on it's way again.

      sheesh

      --
      THE MAGIC WORDS ARE SQUEAMISH OSSIFRAGE
    31. Re:What's in it for desktop users? by jusdisgi · · Score: 1

      But w/ a 100Gb Ethernet you can download the Real Doll data of the supermodel to your desktop fabber real fast!

      Sorry to burst your bubble, but Weird Science was a fictional work...

      --
      Given a choice between free speech and free beer, most people will take the beer.
    32. Re:What's in it for desktop users? by Vancorps · · Score: 1

      Ahhh, Dell D800 here with integrated gigabit NIC. Single transfers of large web logs so it's very easy to transport.

    33. Re:What's in it for desktop users? by bill_kress · · Score: 1

      Like I said, meh.

    34. Re:What's in it for desktop users? by Vancorps · · Score: 1

      What experience is it that tells you throughput for a gigabit link would be that low? Even 100meg with a PCI bus can net you 70meg throughput depending on contents. My experience with gigabit links on my desktop are significantly higher than that. A gigabit link with a PCI bus tops out around 400megabit last I recall. When you move up to PCI-E and PCI-X then you can hit around 800megabit on a fiber connection losing the end with regular tcp/ip overhead. I can do 136megabit using my laptop hard drive to transfer a large web log to my server. That's a crappy laptop with a slow hard drive let alone a modern SATA 150 or SATA 300 hard drive.

    35. Re:What's in it for desktop users? by evilviper · · Score: 1
      As it is, your average desktop will not handle anything even close to 100G Ethernet.

      AND? What would you expect the average desktop users to WANT 100G networking for? So they can fill-up their hard drive in 7 seconds (far faster than the drive can write)?

      At that point, your bottleneck is the PCI or PCI-X bus.

      Well then, thank goodness PCI-express is becomming quite popular on newish systems.

      As the bus has been one of the slowest PC components to innovate,

      See above.

      PCI was significantly over-engineered (and AGP took over the task of graphics), so upgrades weren't needed for a very long time.
      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
    36. Re:What's in it for desktop users? by jibjibjib · · Score: 1
      ADSL2+ is becoming obsolete... Tell that to Telstra. (Australia's largest ISP, started selling ADSL2+ for the first time less than a month ago.)

      http://en.wikipedia.org/wiki/Telstra

    37. Re:What's in it for desktop users? by will.murnane · · Score: 1

      Pci express is 2.5 gbit/s signaling rate per lane with 8/10b coding. So 16x pci-e is 2.5*10/8*16=32 gbps of user data. 100 > 32, last I checked. Adding packet headers and so forth on 100GBe still won't make it three times slower. And since 16x pci-e is the fastest thing going that I know of, this probably works only with switches. From there you demultiplex out to multiple 10GBes.

    38. Re:What's in it for desktop users? by Znork · · Score: 1

      Dude, take a look at PCIe. PCIe is itself layered with physical layer, packet based data layer (including CRC's and packet numbers) and a transaction layer on top of that. It aint that far from ethernet as it is. Heck, take a look at how latency on PCIe is solved today, with credit buffers to avoid waiting for acks from devices...

    39. Re:What's in it for desktop users? by Znork · · Score: 1

      "Pretty sure the PCIe bus can supply 1GByte/s to a 10Gbit card."

      Depends on the number of lanes. Each PCIe lane is about 250MB/s, so you'd need at least a 4 lane slot/card.

      Ordinary PCI, including its 66MHz and 64bit bastard children, which has a fair installed base in the server space and which I suspect the grandparent meant, tops out at around 500MB/s. And with the severe drawbacks of the bus downgrading to the least common denominator, it's not exactly certain you'll actually get anything close to that.

    40. Re:What's in it for desktop users? by Com2Kid · · Score: 1

      LOL, and possibly compressed in transit, making the overall transfer speed seem even higher. :)

      --
      Need help treating your acne? Come here!
  3. Thats 100 gigs, per second? nice by 192939495969798999 · · Score: 1, Funny

    And now for Slashdot madlibs: it would only take a few ______ (large time intervals ) at that speed to backup the average _____ (insert rival group here)'ers pr0n collection!

    --
    stuff |
  4. I prefer Bill Watkins' take on it. by Tackhead · · Score: 4, Insightful
    "The need for 100G Ethernet is growing as IP video and transaction-intensive Web 2.0 applications are exploding across the Internet. Companies such as YouTube regularly add 10Gbps service pipes to meet growing demand, and carriers will need a better way to aggregate such links, industry watchers say."

    - From TFA.

    Which is all well and good, but for honesty, I prefer Bill Watkins' take on it.

    "Let's face it, we're not changing the world. We're building a product that helps people buy more crap - and watch porn."

    Bill watkins, CEO of Seagate

  5. don't trust such initiatives by SilentGhost · · Score: 5, Insightful

    328 feet - it's a good standard, but I like 100 metres better.

    --
    root of all...
    1. Re:don't trust such initiatives by ichigo+2.0 · · Score: 1

      Insightful?

      While I do prefer the metric system, it would be nice if the mods remembered to have their funny detectors on. =)

    2. Re:don't trust such initiatives by Lord_Slepnir · · Score: 1

      Interesting. I was wondering where they pulled that spec out of. My next question is: If they based this one off of metric, why did they base the other one off of Old English?

    3. Re:don't trust such initiatives by SilentGhost · · Score: 1

      I'm sorry, but my comment wasn't meant to be funny. if any international body (or standardizing for that matter) would prefer to use in their everyday practice system not familiar to a vast majority of world's population, I wonder what kind of "standard" will it produce.

      --
      root of all...
    4. Re:don't trust such initiatives by tool462 · · Score: 4, Informative

      Pause for a moment and realize that 6 miles is approximately 10km, which is probably the "real" spec. Not only is it metric, but it's a power of 10, which gives me lots of warm fuzzies.

    5. Re:don't trust such initiatives by slcdb · · Score: 3, Informative

      The actual standard will state the maximum distance in meters, not in feet. Whoever wrote the article did the conversion from meters to feet.

      Just search IEEE 802.3 for yourself. You'll find no mention of "feet". Everything in there is measured in meters.

      --
      Despite what EULAs say, most software is sold, not licensed.
    6. Re:don't trust such initiatives by jez9999 · · Score: 1

      Oh tell me about it.

      I have no idea how people can use imperial measurements. All we have to say is '1 litre'; they have to somehow remember 2.11337641 pints!!

      --
      == Jez ==
      Do you miss Firefox? Try Pale Moon.
    7. Re:don't trust such initiatives by sasdrtx · · Score: 1

      By gum, I ain't using it until its maximum distance is a multiple of 100 good ol' American FEET!

      --
      Most people don't even think inside the box.
    8. Re:don't trust such initiatives by trb · · Score: 3, Funny
      I have no idea how people can use imperial measurements. All we have to say is '1 litre'; they have to somehow remember 2.11337641 pints!!

      And they say 1 pint, and you have to remember 0.47317 litres. Tag, you're it.

      28 g of prevention is worth .454 kg of cure. Ick!

    9. Re:don't trust such initiatives by cyclomedia · · Score: 1

      here in europe that's 0.568 litres, you insensitive clod.

      --
      If you don't risk failure you don't risk success.
  6. Uplink by GigsVT · · Score: 4, Interesting

    What I really want to see is higher uplink ports on SMB hardware.

    Right now, if I want to make a medium size network using lower cost components, it might look something like 5- 24 port, 100-meg switches with 1 GB uplink to a big GB switch.

    The bottleneck here is those uplinks. Each 100meg switch has plenty of backplane, and so does the gigabit switch, but those 100 meg 24 port switches have to share 1GB each to the backbone MDF.

    So I really don't care about PCs or network cards or whatever, just give me 10GB links that I can use between switches without having to pay for overpriced Cisco crap.

    --
    I've had enough abrasive sigs. Kittens are cute and fuzzy.
    1. Re:Uplink by !ramirez · · Score: 1

      Oversubscription models have not ever crossed your mind, have they?

      1:2.4 oversubscription isn't bad at all. Do you really think that a 24pt 100mbps switch needs 10 GIGABIT uplinks in order to work well? If so, I'm sure that Extreme or Force10 would love to sell you some hardware.

    2. Re:Uplink by GigsVT · · Score: 1

      I know, in hindsight my example wasn't too strong, I should have said 48 port switches, or pure gigabit.

      But consider a pure Gigabit network. Right now you'd have to have gigabit IDFs with gigabit uplinks to gigabit MDFs... that's 24:1 oversubscription with 24 port switches.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    3. Re:Uplink by pacman+on+prozac · · Score: 2, Interesting

      You could use something supporting etherchannel and bond a few 1GB links together. We use that to great success, admittedly with using Cisco kit but there's plenty of other companies around making kit that supports channel bonding.

      Incidently what are your users doing that maxes out gig uplinks? We have 96 ports sharing 2x1gig uplinks all over the office without problem, but none are particularly heavy traffic users.

    4. Re:Uplink by Amouth · · Score: 3, Insightful

      just wanted to note.. while yes Cisco is overpriced they most certainly are not crap.. they do exactly what they are supposed to do and they do it well. if you are looking for something that is above the average Joe's network you are going to have to pay whether it be Cisco, foundry, or anyone else - most people that have Cisco switches don't use half the features that they get with them.. they just plug them in and run.. it is their configurability that makes them rock.

      --
      '...if only "Jumping to a Conclusion" was an event in the Olympics.'
    5. Re:Uplink by !ramirez · · Score: 1

      Yes, in which case, if you want pure gigabit performance, YOU HAVE TO PAY FOR IT.

      Can you legitimately justify to me, or anyone else, that an SMB network needs 1gbps access-layer switches, with 10gbps uplinks to distro/core layer switches? If that's the case, then I'll show you a network that needs to be running on something like Cisco, Extreme, or Foundry, and NOT your 'lowpriced SMB switches'.

      You can't build a sports car out of turds and baling wire. Well, you could, but you shouldn't expect much from it.

    6. Re:Uplink by silas_moeckel · · Score: 1

      Channel bonding is your friend. Assuming a top of rack 48-52 port gig switch you can take your 1:48 over subscription down to 1:12 or 1:6. Depending on what your doing a 48+4 switch gets you a nice 1:12 over subscription and 1:24 with a failure assuming you can split up the vlans.

      --
      No sir I dont like it.
    7. Re:Uplink by jabuzz · · Score: 1

      Maybe, but at the moment there is no way on earth we could take 1Gb to the edge at work. The network is two big. Think up to 1400 outlets a patch room, think over a dozen patch rooms in the building. We are not using cheap SMB switches either, it's a combination of managed enterprise HP Procurve and Cisco.

      At the moment it is 100Mbps to the edge, 1Gb uplinks in the patch room, and 1Gb (sometimes two) uplinks out the patch room. We really need 10Gb uplinks out the patch rooms just to get the performance levels the users are crying out for. We really could use 1Gb to the edge, but this would need 10Gb concentrators in the patch rooms, and 100Gb uplinks, and all *way* two expensive. A 24 port Procurve 2900 will set you back over 1000GBP, and some of the patch rooms have over 50 Procurve 2626's in at the moment.

    8. Re:Uplink by GigsVT · · Score: 1

      I said with reasonably priced hardware for a small to medium business. We aren't talking vlans here.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    9. Re:Uplink by GigsVT · · Score: 1

      it is their configurability that makes them rock.

      You mispelled suck.

      We have several Cisco PIX doing routing and VPN. They break all the fucking time, whenever anyone touches anything. Fragile as hell, and hard to debug.

      Our Cisco catalyst switch was the first switch I've ever seen that just crashed completely. We had to reboot it to fix it. Cisco wouldn't even believe us when we told them what happened.

      I've had my fill of Cisco crap. Just because it costs a lot doesn't mean it's good. Look up cognitive dissonance.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    10. Re:Uplink by GigsVT · · Score: 1

      There's no reason we need to have unnecessary bottlenecks on our uplink ports. 10G ethernet has been out of years, this waiting game is stupid.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    11. Re:Uplink by !ramirez · · Score: 1

      You're arguing a ridiculous point - that there should never be oversubscription in a network, anywhere.

    12. Re:Uplink by !ramirez · · Score: 1

      There are full-rack chassis capable of terminating 1200+ 1gbps links, and can support multiple 10gbps uplinks. Specifically, the Foundry NetIron MLX-32. I will be blunt - if you've got 50 24-port switches in a patch room, I can geniunely state that your network planning skills are atrocious. There *are* better solutions out there. You may not LIKE the cost, but to be honest, unless you're ready to build your own switch, it's always a toss-up between cost and features. If you can't spend to get the features, you shuoldn't expect them.

    13. Re:Uplink by Amouth · · Score: 1

      not to be a jerk but if you had a PIX doing routing i can see why it failed, that isn't what it is for. as for the switch what type.. and what do you mean by crash.. i know that just because something is expensive doesn't make it good.. but i have never had any problems with cisco equipment (except refurbs...) now i have seen alot of times when someone put too small of a router or switch in for their needs and didn't know any better.

      and as for Cisco's VPN concentrators.. they are very good if you use cisco's client.. they don't play well with others.. but the routers and switchs are very nice.

      --
      '...if only "Jumping to a Conclusion" was an event in the Olympics.'
    14. Re:Uplink by c_g_hills · · Score: 1

      There's the 3Com 5500G which supports 48x 1Gbs access and 2x 10Gbs uplink ports per 1u switch. They stack up to 8 units to allow up to 448 ports with 96Gbs stacking bandwidth. Individually each switch has 232Gbs switching capacity.

    15. Re:Uplink by silas_moeckel · · Score: 1

      Ah so all dumb unmanaged switches? Small business I would hope, even my low end clients are running hp procurve or similar gear (lifetime maint is hard to beat) 24 +2 port fast e is a few hundred or just over $10 per port. If you really need speed I would think you would be running large MTU and cut through rather than store and forward switching.

      --
      No sir I dont like it.
    16. Re:Uplink by WuphonsReach · · Score: 2, Interesting

      If you don't need VLANs or managed switches, try the "smart" switches which fill the niche between completely unmanaged and fully managed switches. Then you can get things like a 48-port gigabit smart switch for around $1500. Which gives you 40 ports for end-users and 8-ports for uplink or backbone use. Even some of the less expensive "smart" switches support VLANs, but you have to configure using a web browser.

      They aren't the fastest things in the world, but at least they do trunking.

      (Heck, I have a 16-port SMC smart gigabit switch that I picked up for around $240. Gigabit is definitely within reach.)

      --
      Wolde you bothe eate your cake, and have your cake?
    17. Re:Uplink by jabuzz · · Score: 1

      I didn't plan the network and yes the cost is one issue. Terminating 1200 1Gbps links into one chassis is another issue. From a practical perspective it is a nightmare. We don't yet have the money for even one 10Gb uplink, let alone a full chassis switch. Besides a NetIron MLX-32 will only provide 640 ports in a chassis and 1280 in a rack. As I said originally we have just shy of 1400 ports in one room. That forms a small part of the building and an even smaller part of the overall network.

    18. Re:Uplink by GigsVT · · Score: 1

      On a LAN? Why not? There's no reason for it. We have 10 gig technology.

      Why would anyone ever purchase a huge catalyst that has gigs and gigs of backplane bandwidth if there wasn't demand for a scalable way to avoid overscription?

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    19. Re:Uplink by GigsVT · · Score: 1

      This switch was overkill. It was a huge Catalyst that cost way too much. I replaced it with 5 netgear 48 port switches and everything's been fine since.

      What the hell is a PIX for, if it's not for routing? I seriously don't know, I just use whatever Cisco crap the corporate office sends down for routing the VPN.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    20. Re:Uplink by GigsVT · · Score: 1

      Jumbo frame doesn't work well on a heterogeneous network.

      Not necessarily all dumb switches, just not the highest end stuff.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    21. Re:Uplink by Amouth · · Score: 1

      a PIX box is a security appliance - basicly a firewall on so much crak that it doesn't let jack through.. PIX boxes arn't all that wonderful, but they are not routers they do not route anything. http://www.cisco.com/en/US/products/hw/vpndevc/ps2 030/

      as for the switch.. if it died i would have gotten it replaced.. that is what smartnet is for.. the fact that you replaced it with 5 netgear 48's means that more than likly you did have over kill on the switch (if you had one doing that job) and it also means you more than likly where not using it for the fetures. for basic networks yea cisco isn't the answer.. but when you need to truly manage a large network - they have the best tool set.

      I am not saying cisco is the only way to go.. i just don't like when people just call them crap.

      --
      '...if only "Jumping to a Conclusion" was an event in the Olympics.'
    22. Re:Uplink by GigsVT · · Score: 1

      Our PIX is indeed routing, and doing VPN.

      So they can route, apparently just not very well.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    23. Re:Uplink by Amouth · · Score: 1

      what model do you have.. i have never seen one that could route..

      --
      '...if only "Jumping to a Conclusion" was an event in the Olympics.'
  7. Umm.. hey..? by SuperStretchy · · Score: 1

    Hey, if I help out.... Can I get it for free? My imaginary OC-192 is getting a tad slow and my imaginary income won't allow me to feed my imaginary family and get a 2nd 192.

  8. Its already done by warrior_s · · Score: 3, Interesting

    Not exactly but Bell Labs did something like this in March http://www.lucent.com/press/0306/060308.coi.html

    1. Re:Its already done by warrior_s · · Score: 1

      Sorry for the incorrect link. I didn't realize alcatel and lucent merged and I simple copied the link from my bookmarks. Here it is again 100G

  9. Funny you should mention... by Majik+Sheff · · Score: 1

    This sounds like an excellent technology to adapt for large cluster interlinks. It'd be nice to have insanely large pipes that go further than a few meters before the bits spill out.

    --
    Women are like electronics: you don't know how damaged they are until you try to turn them on.
  10. Re:Imagine.. by smoker2 · · Score: 1

    Doh !

  11. Ethernet speed vs. PCI/PCI-X/PCIe speds by this+great+guy · · Score: 2, Informative

    You are right. Here are some numbers for the curious, nothing comes close to 100 Gbit/s:

    PCIe x16 (2.5 Gbit/s per lane, 8B/10B encoding): 32.0 Gbit/s bidirectional (64.0 Gbit/s of aggregated bandwidth)
    PCIe x8 (2.5 Gbit/s per lane, 8B/10B encoding): 16.0 Gbit/s bidirectional (32.0 Gbit/s of aggregated bandwidth)
    PCIe x4 (2.5 Gbit/s per lane, 8B/10B encoding): 8.0 Gbit/s bidirectional (16.0 Gbit/s of aggregated bandwidth)
    PCIe x1 (2.5 Gbit/s per lane, 8B/10B encoding): 2.0 Gbit/s bidirectional (4.0 Gbit/s of aggregated bandwidth)
    PCI-X 2.0, 533 MHz, 64-bit: 34.13 Gbit/s
    PCI-X 2.0, 266 MHz, 64-bit: 17.07 Gbit/s
    PCI-X, 133 MHz, 64-bit: 8.53 Gbit/s
    PCI, 66 MHz, 64-bit: 4.27 Gbit/s
    PCI, 66 MHz, 32-bit: 2.13 Gbit/s
    PCI, 33 MHz, 32-bit: 1.06 Gbit/s

    However, regarding 10G ethernet adapters, does anyone know when vendors will start making use of PCIe x8 or x16 for them ? In all those Internet2 benchmarks papers, everybody complains about PCI-X beeing too slow, but PCIe x8 or x16 would be perfect for 10G.

    1. Re:Ethernet speed vs. PCI/PCI-X/PCIe speds by jabuzz · · Score: 1

      You mean like Myricom

      http://www.myri.com/Myri-10G/10gbe_solutions.html

      Not bad prices, list is 795USD for a fibre optic card and 500USD for a SR optic or 900USD for a LR optic. A CX4 card is only 695USD. With switches like the HP Procurve 2900 having 10GbE CX4 as standard, I predict that 2007 is the year when 10GbE really moves mainstream.

    2. Re:Ethernet speed vs. PCI/PCI-X/PCIe speds by this+great+guy · · Score: 1

      Wow, thanks for the info. I had never read about the release of this 10G PCIe NIC !

  12. Re:Imagine.. by DittoBox · · Score: 1

    I for one welcome our nerdy hackneyed-joke telling overlords.

    Ahh...screw it.

    --
    Good. Cheap. Fast. Pick Two.
  13. Re:Not for home use... by popeye44 · · Score: 1

    We consistently use 4gb tiff files where I work, While compression technologies are great.. we tend to saturate our lines on busy projects. We have 2 Ds3's. More is always gooder :D

    --
    Inane Comments are Generously Disregarded
  14. *sigh* by M0b1u5 · · Score: 4, Insightful

    Strange. The standard is to be "six miles" over single mode, and "328 feet" over multimode.

    I don't get it!

    I mean, we KNOW all decent standards use metric measurements - and Americans are inclined to convert them to the National Stupid System, so 328 feet makes sense (100 metres) - but where does this "6 miles" business come from? It is only 9,660 metres (9.66 km).

    Surely the standard will be 10,000 metres - ten kilometres, and the poster was lazy, and couldn't be bothered with the extra 0.2 of a mile?

    My question is this: when the specification is clearly based on very simple numbers: 100 metres and 10,000 metres - why convert that into the Stupid System? /.ers are not so stupid as to have to be fed figures fudged for obscurity!

    --
    How many escape pods are there? "NONE,SIR!" You counted them? "TWICE, SIR!"
    1. Re:*sigh* by businessnerd · · Score: 1
      Don't blame the poster, blame the fucking article.

      From the fucking article:
      The IEEE will work to standardize 100G Ethernet over distances as far as 6 miles over single-mode fiber optic cabling and 328 feet over multimode fiber.
      Also Network World is based in the United States, so it can be assumed that the majority of their readers are also United Statesmen, and therefore, would recognize miles and feet more quickly than kilometers and meters. As for the rounding, I could really care less if 6 miles is not exactly 10km. There close enough, I get the idea.
      --
      "It's not whether you win or lose, it's how drunk you get." -- H. J. Simpson
    2. Re:*sigh* by Grave · · Score: 1

      But how many hogs heads can I get out of that?

      Damn kids and their "standard" measurements. In my day, we measured the speed of the connection by how many beers we could drink before the nipples were visible!

    3. Re:*sigh* by omry_y · · Score: 1

      News flash:
      Meters, miles, feet etc are just measuring systems.
      sometimes the nasty laws of physics gets in the way of the divine goal to work with round nice numbers.

      --
      Omry.
  15. Re:Not for home use... by Zantetsuken · · Score: 1

    where do you work? a map-making company or professional high-res (REALLY high) photography?

  16. More units stupidity in the article by Beryllium+Sphere(tm) · · Score: 1

    >push Ethernet to a megabits-per-second speed that does not currently exist under any standard

    >a comparable 100Mbps standard does not exist now for Ethernet to emulate,

    And then neglecting the question a journalist should have asked to add value over a press release, namely "Isn't this going to be way more expensive even than FDDI? How many machines have to be talking on the same LAN segment before this gets cost-effective?"

  17. Desktop bottlenecks are likely to go away by nokiator · · Score: 1
    It is true that 100GbE will not target desktop users when it first becomes available. Initially, new flavors of higher speed Ethernet is typically used in switch-to-switch connections. Higher speed links make it easier to aggregate traffic from lower speed links to a single logical link. It usually takes at least 4-6 years for a faster Ethernet standard to propagate from core/distribution applications to server/desktop connectivity.

    However, the current PC architecture is not actually too far from removing the bottlenecks in being able to support a 100GbE interface. A 16-lane PCI-E interface has about 32Gb/s of bandwidth in each direction. The upcoming PCI-E 2.0 doubles the bandwidth per lane, such that a 16-lane PCI-E 2.0 interface would support about 64Gb/s of bandwidth in each direction. When PCI-E 3.0 comes out, it would most likely double the lane bandwidth again and provide sufficient bandwidth to support a 100GbE NIC. Unlike the PCI bus, which was a shared bus, doubling the signal rate on PCI-E may turn out to be easier since PCI-E is a point-to-point signaling architecture.

  18. Re:Little off topic.. by jibjibjib · · Score: 1
    Basically, is it theoretically possible for wireless to eventually achieve 100gb/s speeds?

    Not without taking up a huge chunk of the radio spectrum.

  19. What I was wondering by einhverfr · · Score: 1

    is why the 6 mile limit on single mode fiber. If you can boost the signal in-place (using erbium-doping), why would that limit exist?

    Obviously the limit on multi-mode fiber is understandable though.

    Or am I missing something?

    --

    LedgerSMB: Open source Accounting/ERP
    1. Re:What I was wondering by Ajehals · · Score: 1

      Erbium-doping not withstanding - (I have no real experience with fibre over more than a few thousand meters, and haven't seen the need yet for looking at amplification...), I know that we used to see some signal degradation over distances of 1-2km, with fairly high grade fibre, and even had issues with usability at 800m with some of the lower quality stuff - interestingly this seemed to be more pronounced when using network hardware - in this case HP kit, whilst Telecom related equipment seemed more robust.

      I would assume that there are technical costs associated with increased range and I would guess that those costs would manifest themselves as an impact on any data transmission rate - whether through increased noise / error correction requirements.

    2. Re:What I was wondering by einhverfr · · Score: 1

      What sort of fiber were you uing?

      Quick primer:

      Single-mode (the one with the 6-mile limit) is a strand with the same thickness as the amplitude of the light wave. Therefore the light has only one path down the strand and the signal will only degrade due to attentuation and not "blur" like you see with multi-mode cables. This, along with erbium doping, is what they use in long-haul cables such as those under the Atlantic Ocean. Single-mode can only be used with lasers. LED's won't work. Also single mode is a lot harder to work with (splicing it, for example is extremely difficult).

      Erbium doping in single-mode strands is used to create sections where, when electrified, the strand will act as a laser amplifier (i.e. like a flash tube) and release laser energy only when laser light passes through it.

      Multimode fiber cables are so called because they are thick enough that light can take multiple paths down them. Over distance, you will see signal blurring because what starts out as a discrete flash ends up coming out along a bell curve in terms of time. Multimode can be used with lasers or LED's.

      Even in multimode fiber there are many different grades. The lowest grade is step-indexed which means that the refractive index is such that light bounces off the side.

      If I could guess from your post I would suggest you were probably using grade-indexed multimode fiber. This sort of fiber gradually changes refractive index from the core to the outside, essentially acting as a series of lenses, redirecting the light back to the center. It has quite a bit less signal blurring than step-indexed multimode cables because the difference in the distance between the possible paths the light takes is less, but the signal will still blur.

      --

      LedgerSMB: Open source Accounting/ERP
  20. Found my answer by einhverfr · · Score: 1

    Single mode fiber is still subject to polization-based dispersion and chromatic dispersion. It seems that with cheap lasers, the 6 mile limit might be an issue.

    Note however that erbium doped amplifiers only hit a small portion of the frequency so the dispersion is far less where these are used.

    --

    LedgerSMB: Open source Accounting/ERP