Slashdot Mirror

I don't think you know what a switch does.

Maybe the problem is worse than that! Maybe Cisco themselves don't know what a switch does, since they offer an IDS module for their flagship core switch:

http://www.cisco.com/en/US/pro...

And this seems like a serious problem in the networking industry. Apparently Alcatel-Lucent doesn't know either, since there's a built-in IDS in their core switch.

http://enterprise.alcatel-luce...

And - OMG - even HP is completely confused about this technology, since they also have IDS on their core switch.

Or maybe, just maybe, you vastly overestimate your understanding of enterprise networking.

The Shannon–Hartley theorem (aka Shannon capacity) is the term you're looking for. Modern wireless networks use MIMO (multiple input multiple output) concepts to boost this capacity and squeeze more bits into the same slice of spectrum.

If I take the "single strand" mentioned in TFS very literally, then it is indeed a single strand without amplification.

(Unless you can amplify light on a single strand, without ever breaking it into two or more strands.)

Aha. I went to the Alcatel Lucent site and read The Fine Press Release where the actual truth* was published.

Paris, July 16, 2013 — Alcatel-Lucent (Euronext Paris and NYSE: ALU) has broken a new record for the amount of data that can be transmitted over transoceanic distances on a single optical fiber.

In a test carried out at Alcatel-Lucent’s Innovation City campus in Villarceaux near Paris, researchers from Bell Labs successfully sent data at speeds of 31 Terabits-per-second (Tbps) over 7200km – a capacity exceeding that of the most advanced commercial undersea cables today by a factor of three. This was achieved with a span - the distance between amplifiers maintaining the entire length - of 100km.

The researchers were able to achieve the highest-ever capacity for undersea data transmission on a single fiber.

So there we have at least a few facts from the source. Yes, they used repeaters in a typical undersea configuration. No, they didn't say if these repeaters were the existing erbium doped optical amplifiers or if they used some other novel technology that would require either laying new cables or dredging up the old ones and splicing in new repeaters.

* If you can accept the concept of 'Truth' in a press release.

Alcatel-Lucent has a nice multi-vendor solution for that : http://www.alcatel-lucent.com/products/motive-network-analyzer-fiber

A single 100 Gbps port. http://www3.alcatel-lucent.com/features/100g_era/ is one of the routers that handles 100G. And many others do as well. Then you take the 100G and DWDM 44 of them together for 4.4 Tbps on a single fiber pair. Put that in a 24 pair cable for over 100 Tbps over a single cable. Sell 1/1000th of that to the government and have them pay 99% of the cost for the network, and you have a good profit model.

Here's the index of the July-August 1978 issue where the whole series of articles appears. Better format than the search above.

Several issues of the Bell System Technical Journal tell the story of UNIX, in their own words. This one in particular is interesting.

Several issues of the Bell System Technical Journal tell the story of UNIX, in their own words. This one in particular is interesting.

My understanding is that Bell Labs tested a number of layouts before settling on the 1-2-3 matrix we use now as being simplest to master (see R. L. Deininger, Human Factors Engineering Studies of the Design and Use of Pushhutton Telephone Sets, 1960, Bell System Technical Journal [PDF]).

I'm not sure if calculator / comptometer manufacturers had their competing studies; I've heard that when Bell asked for an explanation, the answer was a shrug...comptometers were about 80 years by then, so I think the origins of their layout are as opaque and full of folk explanations as the QWERTY layout.

Regardless, I've encountered OP's request before...but for phone layouts which matched calculator layouts. I was working in an operations office a few years ago run by a person who was a fan of "Cheaper by the Dozen" who wanted to optimize our phone dialing speed (this was a fun place to work, even if this request sounds odd). We didn't have any success, but it was an interesting thought.

The problem is short attention spans, and a difficulty in communicating the benefits of long-term, fundamental research combined with a political, financial and popular culture obsessed with a "that was yesterday, what have you done for me lately" mentality.

A perfect illustration is shortly after the "merger" of France-based Alcatel and U.S.-based Lucent Technologies was the virtual kneecapping of Bell Labs.

Then CEO Patricia Russo announced that long-term, fundamental research would no longer be performed at Bell Labs as that wasn't the culture of Alcatel. If a project couldn't be productized in 7 years, it would be shelved.

To me that was a "break out the shovels" moment. As in, "It has been a long, hard decline but we can see the bottom. Break out the shovels, we're going to dig this hole deeper."

The same thing goes on with Congress and funding basic scientific research at placed like NASA, the various National Labs (Sandia, Lawrence Livermore, Fermi, Oak Ridge, etc.). Just look at what happened to the Superconducting Super Collider.

The problem is you can't always predict what benefits will come from fundamental research, thus you can't give the bean counters a predicted return on investment number. When is an even harder number.

The only real time the United States as a government priority has pumped money into research is if that research could be used to blow shit up. Actually, this is probably true of Britain, Germany, Russia, Japan and China as well.

We need to be able to clearly articulate the benefits to society and the economy as a whole that fundamental research brings. If we want to drive forward into the future imagined by the visionaries, and not end up in the one envisioned by the dystopians (no Mad Max, please), this and education need to be our top priorities as a nation. Which nation? Any and every nation.

These require fiber backhaul to a baseband processor, so no, they're not really designed for that.

On ALU's website, they say the cube would have microwave for backhaul and could use solar or wind for power, so for those cases it could be used for that.

That was a long time ago. And the important part was the "Bell Labs".

It's still the case at Alcatel-Lucent Bell Laboratories.

Yes, the story headline is talking about something totally different. I mean, how do YOU read "100-Petabit Internet Backbone"? Most people would not interpret it to mean "15.5 Tbps delivered over 7000 km." (The headline error is repeated in TFA. Ironically if you click all the way through to the AlcaLu press release the headline is more accurate: "Alcatel-Lucent Bell Labs announces new optical transmission record and breaks 100 Petabit per second kilometer barrier".)

I will grant you that optics geeks may find the bandwidth-distance metric familiar... but I continue to assert that [Inter]net geeks do not.

They show Tuckerton, NJ near the Va/NC border, and Manahawkin on Virginia's Eastern Shore.

Fascinating stuff. I'm still amazed that we have underwater cables at all. I had be shown a map of existing cables before I believed it. http://www1.alcatel-lucent.com/submarine/refs/index.htm http://networks.cs.ucdavis.edu/~zhuk/maps/alcatel_large.gif

Lucent is a company that does network services and such. Lucid means easily understood.

The ping time only gives you latency. Latency is mostly interesting for online gaming. The difference between 200 and 400ms isn't going to get noticed by anyone surfing and certainly not for big downloads or streaming. Also, cable length has very little influence on latency, probably being the cause of only 10-20% of it. Light goes at almost 300KM/sec and as you point out the AJC is only 12,700KM long. Much of the latency is caused in TCP/IP level switching. If an ISP ordered a circuit from Sydney to LA via AJC and Unity, this circuit would be switched at a much lower level that ads virtually no latency. (and this hop won't show up in your traceroute either) A good example:

  5 pos4-0.bdr1.syd7.internode.on.net (203.16.212.21) 203.835 ms 203.779 ms 203.355 ms
  6 pos2-0.bdr1.sjc2.internode.on.net (203.16.213.41) 202.367 ms 202.518 ms 202.337 ms
  7 ge-6-20.car3.SanJose1.Level3.net (4.71.112.85) 202.347 ms 202.269 ms 202.844 ms

The trace goes straight from Sydney to San Jose, even though the cable passes through Hawaii and possibly Auckland too. It simply is a direct circuit.

We also do not have several independent links to the US already. The only significant link we have are the two loops of the Southern Cross cable, both of which have landing points very close together and both go through Hawaii - an undersea landslide or other plate moment could easily take both out at the same time. The only significant(-ish) backup we have is the AJC, which is already at capacity. (which is only half of SCC to begin with!) It would keep us connected to the world if SCC goes down for sure, but it won't be fast.

This new cable system however WILL give us the high capacity redundant link. With the new 2Tb PIPE cable to Guam (over twice the capacity of SCC and AJC combined) going live next year and Unity's southern loop going through Guam also in 2010 it will give us an enormous boost in capacity and redundancy. It also gives us much more capacity in the other direction around the globe, making it feasible to go to Europe via the shorter route instead of the US. I was a bit surprised about the PIPE cable because there isn't enough connectivity currently on Guam to fill up 2Tb. The announcement of Unity makes it all clear, however.

This is a much better map of undersea cables also.

The image linked from the summary does not depict the physical locations of cables, but is a schematic of existing connections between points on the globe. The lines in that image have not much to do with where the cables actually are. A more realistic representation of (a subset of) the world's submarine cable networks would e.g. be this big PDF or, in a more comprehensive view, that one (sold for a mere $350 :-| ).

Well, I have been testing a mobile product with a similar spectrum (and as a tester, I will here by post anonymously). I have been testing a product in the 800Mhz spectrum, which is in many ways similar. The great thing about 700Mhz is that you are able to achieve a much higher penetration. You are able to provide true mobile broadband that is as mobile as GSM and other cellular technologies. It's even more mobile than Mobile Wimax (when mobile Wimax is defined in the 2.5Mhz spectrum). This enables you to have high-speed internet connectivity while travelling in a car or in a train. As the penetration of the signal is stronger in 700Mhz, you can have fairly large cells, easing the roll-out.

However, the problem is nobody has yet made any indication of what modulation/technology they will be running in 700Mhz. Flarion (now Qualcomm) has Flash-OFDM[1], which is the product I have been testing allows theoreticaly around 5Mbit downstream with 2,5 or so upstream, and is well-suited for 700Mhz, but I doubt that it will be the standard here. One platform has already been specifically built for 700Mhz, and that is a CDMA-2000 platform by Alcatel-Lucent[2]
However, the 700Mhz license only grants usage of about 60Mhz of spectrum, and according to this article, further 24Mhz will be reserved for public safety, and not included in the auction, so how useful is this service anyway, when you are so limited in your channel bandwidth?

Whoops the reference should be:

http://www.alcatel-lucent.com/submarine

Apologies.

This question has come up once or twice before.

The usual suspects for answers to this question are as follows:

NorthStar, which is quite feature rich. "NorthStar is a system to help track and allocate blocks in an IP Network"

IPplan which is another open source product.

And PHPip

If you want to go commercial VitalQIP Enterprise could suit your needs quite well.

Berny

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