For many in the industry commercial 100Gbps can’t get here fast enough, but it is coming! Today at the SC09 conference in Seattle, a coalition of made up of Juniper Networks (NASDAQ:JNPR, news, filings), Infinera (NASDAQ:INFN, news, filings), Level 3 Communications (NYSE:LVLT, news, filings), Internet2, and ESNet demonstrated a multivendor 100Gbps routing and optical network. This group has been working steadily together since last November. In this case of course, we are talking about the 10x10Gb version of 100G, which requires fewer technical advances and will probably come first.
The test involved a new 10x10GbE physical router card from Juniper which they announced today. A 100G test signal was sent from Seattle to Portland and back via Infinera gear over the Internet2 and Level 3 networks – which of course are the same intercity fiber route, but still. Now, this is not a 100G wavelength solution, it is about delivering a 100GbE product to customers using a network with a whole lot of 10G pipes bundled together. But since it doesn’t have to wait for them to work out the kinks in the 100G native wavelength model, it is probably closer to commercial viability.
So when do we see that commercial product then? Who will be the first to offer a 100GbE intercity connection to someone other than a trade show audience? Perhaps we will see something by springtime?
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!
Categories: Internet Backbones · Telecom Equipment
Join Our LinkedIn Group
Rob,
Interested in your view. To me, this cuts both ways– the ever expanding ability to pump more data/p/s over the same pipe continues to dilute the advantage of a very robust network such as Level 3’s which has surplus capacity. maybe those spare conduits will remain an intercontinental mouse transport system into the indefinite future?
I would say that there is still no pressing need for Level 3’s extra conduits, and there won’t be until someone invents fiber that changes the economics. Since it’s out of their control, Level 3 will of course ride the 100G bandwagon wherever it goes, but would probably be happiest if the 10x10G solution is most economical for those with the most fiber available.
Very confused on this fiber principle tied to “useful life.” Does fiber itself, manufactured by Corning as one, not follow Moore’s law? I’ve asked it before while always referencing Jim Crowe’s commentaries which stated this being true, “in the beginning.” I see the carving up or splitting of fiber strands previously extended supply via DWDM. At the same time, you’re pointing to a potential fiber baron’s satisfaction towards a carved up circuitry solution rather than one fall swoop at addressing the equipment need, i.e., less equipment being more cost effective. I assume that makes it less cost effective, inuring to (3)’s longer term benefit, if I am understanding you correctly. I guess the question is, when does “existing fiber” in existing conduits representing Mother Earth’s current supply, become obsolete, or unusable in the marketplace, while splitting and carving, and jump starting it with all this new equipment remains the status quo?
When it comes to the useful life of fiber, think of the nine lives of cats. Evey time it appears that one of its properties has reached its elastic limit (think: polarization mode dispersion beyond 10Gbps), another fix somehow finds its way into the mix to once again quadruple yield. Thus far only on-off (RZ, NRZ) signaling and some latter day variants of quadrature modulation schemes have been used. When things start looking really desperate (again), and it appears that once again fiber’s potential is nearing exhaust, I’ll be watching for a return to down-and-dirty, full-blown AMFM and other sub-band and sideband modulation techniques, a la cable tv, superimposed onto each of the then already-DWMD’ed lambdas being employed. They say that it takes more than a millennium of gravitational force to make glass begin to sag and droop. Barring any ill effects from radiation and excessive temperature changes, most fiber currently in the ground should still be good to go, with appropriate forms of modulation, at least, for some time to come. Leastwise, long after it will make much of a difference to anyone reading this. Whether the marketing machines of the glass manufacturers will be able to convince operators otherwise, however, is another story entirely.
For the most part, I have to agree with you. However, with certain generations of fiber, there have been noted in the field a serious degradation in the ability of the fiber to transmit. From what I understand, this has been due to some problem in the outside coatings of the individual fibers that has hampered the internal reflectivity of the fiber. Essentially what happens is the signal “leaks” out of the fibers over a relatively short distance.
Good point, Clevus. Thanks. I should have also noted the vulnerability to fiber in high-stress situations, water infiltration, fractures caused by micro-bends and fatigue, etc. Interestingly, during the Q&A session at the end of today’s broadband.gov hearing in Washington http://bit.ly/4j6TG5 , Verizon’s Dick Lynch and ECFiber’s Tim Nulty spent several minutes on this topic, stating essentially what I’d stated earlier. Not sure if it’s scheduled for replay anytime soon. For what it’s worth, I found this hearing to be time well spent.
As relates to the useful life of fiber, another qualifier to my original comment above:
Of course, degradation of existing fibers wouldn’t be the only cause for change. If a grade of fiber should happen to come along that was eminently more permissive than today’s singlemode fibers, then there would be cause to rethink the need for upgrading. While not in the cards for the immediate future, I keep reading about POF (polymer optical fiber) advancements, and more interestingly now, about PFC (photonic crystal fiber) that may someday steal the spotlight away from today’s silica based strands.
See: Photonic-crystal fibres: the route to success
http://optics.org/cws/article/research/40763
Frank, you sir, are another very sharp tool in this blog’s tool shed.
Based upon what you’ve stated, I do have another question.
Because you are very verbose and technically gifted, I might add, I must confine you to a yes or no answer. 🙂
Was the (3) Brain Trust, including Gates whispering in the back ground, wrong for incurring such great expense on the front end when “millenniums” are apparently the useful life?
IMO, yes. Given the national footprint that was initially decided, then yes, there was wisdom in incurring the marginal costs related to the additional ducts, if that is your question. While I’ll refrain from commenting about the “gifted” comment in your reply (except to say thanks, I think) I’m rather certain that Rob will agree with you about the verbose part 😉
oops… see what happens when I’m tied to a single word answer? Yes they were wise, is what I meant, i.e., no they were not wrong.
I think I’m going to call Jim Sinclair, since giving up on Steve Ballmer. (3) needs a significant investor to gain credibility in the market place again, since they’ve already been through the toxic PIPE’s by scoundrels and are clearly ready to PRODUCE BROADBAND GOLD in QUANTITY tied to very high margins. Jim understands the front loaded bucket now behind us-capital intensity- of approx. $25B in capital investments made the past eleven years because, he was involved with national cable company investments. Longleaf and Fairfax remain illusive at giving this security credibility. Some of their interest is pure usury, nothing less than TAKING by TAKERS to me.
I’m with Buffett on shiny metals being of less utility or value in today’s world compared to his own father’s desires, possibly Sinclair’s too, by coveting a gold standard versus the “cash flows” of quality fixed or hard assets with business moats sunk into the ground!
http://www.kingworldnews.com/kingworldnews/Broadcast/Entries/2009/11/13_Jim_Sinclair.html