Xtera Communications Inc. raised eyebrows by announcing a second funding round of $110 million yesterday, bringing the total it's raised to $133 million.
"I'm surprised. They seemed quiet lately," says Scott Clavenna, president of PointEast Research LLC and director of research at Light Reading. "I don't hear much about them. It seemed as though they were either getting in trouble or getting their act together."
Xtera says the latter's the case, and that it's been able to attract investors by solidifying its value proposition. That includes a new product slated for delivery in March, a so-called "ultra-wide" amplifier for DWDM (dense wavelength-division multiplexing) systems that Xtera claims will open up 96 extra 10-Gbit/s channels by tapping into a little-used portion of the optical spectrum (1480 to 1510 nanometers). Carriers who buy the amplifier, Xtera says, will be able to literally double the amount of bandwidth they offer without lighting new fiber.
The amplifier will work within the distance limits of most DWDM gear today, Xtera says. In the future, though, it plans to also offer ultralong-haul subsystems that would reach in excess of 1,000 kilometers.
There's another surprise hidden beneath these claims. Xtera's new products are designed to be tacked onto existing DWDM kit from vendors such as Ciena Corp. (Nasdaq: CIEN). That's a different market approach than Xtera was giving out a few months back (see Xtera Communications Inc.). At that point, the vendor seemed ready to compete with a whole gaggle of vendors and startups developing ultralong-haul DWDM systems (see The Ultimate Backbone ).
Now, it looks as though Xtera's found a preferred identity as a partner to the big DWDM players. "We're neither a component nor a system vendor," asserts Paul Harrison, vice president for marketing.
Was this approach the selling point for Xtera's high rollers? It's tough to tell. Some sources think the influx of cash, while large, is only to be expected, given current market conditions. "Frankly, I think it shows life in the market," says Seth Spalding, director at Epoch Partners. "We've seen these kinds of numbers before, but chiefly through public financing in IPOs." He says that as the public market backs off on IPOs, the private sector is starting to increase its level of funding.
"Traditionally, the private sector's taken on greater levels of risk. Some companies that may have transferred to the public
market in the recent past are now deciding to raise another private round of funding instead," Spalding says.
re: Xtera's $110M Surprise Xtera makes a case for adding value to a DWDM product that a carrier may already have purchased. Seems likely to save costs and hassles. Any downsides here?
re: Xtera's $110M Surprise It's a nice idea, but you can't just add bandwidth like that.
There are two main technologies for the 1450-1510 region, Thulium and Raman. Both currently suffer from poorer performance than the erbium amplifiers that are used today.
1) the S band amplifers are noisier, therefore you will need to place them nore often than the current erbium amps. But your huts are at fixed locations - figure that one.
2) stimulated Raman scattering (SRS) is a non-linear effect that affects ultra wide systems. It causes power to be robbed from short wavelength channels and fed into the long wavelength channels. So if your system was not designed day one for S band, you won't be able to upgrade.
3) Ciena and others have highly proprietary equipment management software. To integrate into that system will be a difficult job at best.
So it will not save costs - it will be cheaper to light a second fibre. Do Xtera think that Ciena etc are not going offer S band?
re: Xtera's $110M Surprise Every carrier I've heard speak at recent conferences has said they want to get as much out of the existing lit fiber before lighting the next. Sounds like a great proposal to me
re: Xtera's $110M Surprise I don't think a company raises $110M if it hasn't solved those problems. No-one's going to pay more $/Gb/km so I'm sure they think they've cracked that. As to Raman amps recent papers have shown that Raman amps can get upto 40dB gain, even more than EDFA's so the same span designs should be no problem. Thulium doesn't dope into Silica fibers so you cannot fusion splice Thulium doped amps to line fiber or other silica pigtails from other amp components making them practically useless!
re: Xtera's $110M Surprise 1) the S band amplifers are noisier, therefore you will need to place them nore often than the current erbium amps. But your huts are at fixed locations - figure that one.
Raman amplification is noisier?
2) stimulated Raman scattering (SRS) is a non-linear effect that affects ultra wide systems. It causes power to be robbed from short wavelength channels and fed into the long wavelength channels. So if your system was not designed day one for S band, you won't be able to upgrade.
From the Xtera web site: "When upgrading a C-Band system to include the S-Band, the C-Band will receive power transfer from the S-Band. To adjust for this, the C-Band will need some downward adjustment of power/gain. Downward adjustments are much simpler operations than upward adjustments on systems already set to maximum power. Since the wavelengths in the S-band are lower than those in the C and L-Bands, Xtera's S-Band amplifier has been designed upfront to accommodate for this effect."
3) Ciena and others have highly proprietary equipment management software. To integrate into that system will be a difficult job at best.
Probably won't do it. Treated as separate managed elements who happen to be sharing the same fiber. Isn't that how Nortel gets to the terabit range with what ostensibly is 'one product'??
So it will not save costs - it will be cheaper to light a second fibre.
After all is said and done, it may be the case.
Do Xtera think that Ciena etc are not going offer S band?
re: Xtera's $110M Surprise "1) the S band amplifers are noisier, therefore you will need to place them nore often than the current erbium amps. But your huts are at fixed locations - figure that one.
Raman amplification is noisier?"
Lumped Raman is noiser than a lumped erbium amplifier. Forward pump the Raman and you get pump noise transfer, backward pump it and you get imcomplete saturation. Both mean than it is noisier.
Thulium IS soluble in silicates, just not in silica. Take a look at OFC if you don't believe me.
2) "From the Xtera web site: "When upgrading a C-Band system to include the S-Band, the C-Band will receive power transfer from the S-Band. To adjust for this, the C-Band will need some downward adjustment of power/gain. Downward adjustments are much simpler operations than upward adjustments on systems already set to maximum power. Since the wavelengths in the S-band are lower than those in the C and L-Bands, Xtera's S-Band amplifier has been designed upfront to accommodate for this effect."
Oh Good. So current transmission kit has tilt control as well as power control. NOT. If you turn down the pumps on a C band amp, the short wavelength end drops further than the long wavelength end. SRS boost the long wavlength end more than the short wavelength end. So your C band will end up extremely tilted. And so your system won't work. Let me guess, Xtera will add some filters to correct this. Now that will really screw your link budgets.
re: Xtera's $110M Surprise Regarding point 2), one can always use an LBO (some fixed attenuator) to lower the output power of a C-band amp *without* causing gain tilt.
I doubt that the C-band signals will experience *any* Raman gain at all, since the overall "S plus C" band amp must be a split-band design. Otherwise the 1480 nm part of the signal will be absorbed by the C-band EDFA.
re: Xtera's $110M Surprise "I doubt that the C-band signals will experience *any* Raman gain at all, since the overall "S plus C" band amp must be a split-band design. Otherwise the 1480 nm part of the signal will be absorbed by the C-band EDFA."
Stimulated Raman Scattering (SRS) occurs in the the transmission fibre itself, and can cause up to 6 dB of tilt. Unfortunately, the tilt is the same direction as the EDFA gain tilt.
You can't just add an attenuator to an EDFA without altering the link budget. Nor is there any way of accessing it. You will probably need to throw away the C band amp - making the cost model dubious.
Add on the fact that the fibre has more loss in the 1450nm region (whether there is water or not), so you will need new link budgets.
Retrofitting systems is a tricky prospect. Do you really think that Ciena and Nortel will stand idly by and not offer their own upgrades if it was economically sensible?
re: Xtera's $110M Surprise Looking at the Xtera web site they plan to sell the technology via Nortel, Ciena etc as they have it available and the internal developers do not. This is not a competing product to lay along side an existing system in the field, it becomes part of it, managed by it and supported by the services providers existing vendor
To save this item to your list of favorite Light Reading content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
(17) | 
Digg
Del.icio.us
Reddit
Tweet This