AlcaLu, Ciena Look Good for Verizon RFP

Verizon Communications Inc. (NYSE: VZ) has issued its request for proposals (RFP) for a new packet-optical transport system (P-OTS), but as expected, the system that best matches the specs is almost certainly out of the running.

That's because it comes from Huawei Technologies Co. Ltd. , writes analyst Michael Genovese of Soleil Securities Group Inc. in a note issued this week.

Verizon's RFP details what it calls a long-haul optical transport platform (OTP). It's part of a planned optical architecture that uses reconfigurable optical add-drop mulitplexers (ROADMs) and the Optical Transport Network (OTN) standard for switching, saving core routers for the traffic that really needs routing. Verizon officials have publicly talked about the idea since March. (See Verizon Rethinks Long Haul and Packet Optical Transport Goes for the Long Haul.)

Genovese and Heavy Reading analyst Sterling Perrin presented an overview of the optical market to Soleil's customers recently, with the Verizon RFP as a headline topic. Their conclusion is that once it starts shipping a 100Gbit/s interface, Huawei's OSN 8800 (a Leading Lights finalist) will have the closest fit to Verizon's specs. Perrin had come to a similar conclusion in May. (See Vendors Target the Packet-Optical Core.)

One tiny problem: "We believe Huawei has almost no chance of winning a major Verizon deal," Genovese notes.

While he doesn't specify the reasons, it's no secret that US government officials -- and, therefore, AT&T Inc. (NYSE: T) and Verizon -- are reticent to put packet traffic on Huawei gear. (See US Gets Worried About Huawei .)

Based on what the Perrin/Genovese tag team is saying, here are some requirements for Verizon's OTP:
  • It handles 80 channels of 100Gbit/s long-haul DWDM.
  • It uses a multi-terabit Ethernet switching fabric.
  • Wavelengths can go more than 1,000 km without regeneration.
  • Sonet/SDH traffic gets handled by the Optical Transport Network (OTN) standard, with grooming down to the ODU-0 level (equivalent to 1 Gbit/s).
  • The control plane must support a mesh architecture and run on standards such as the Automatically Switched Optical Network (ASON) and Generalized Multiprotocol Label Switching (GMPLS).

So, what fits the bill, if not Huawei? Genovese and Perrin like the Alcatel-Lucent (NYSE: ALU) 1870 Transport Tera Switch and the Ciena Corp. (NYSE: CIEN) 5400. (See AlcaLu Joins the War for the Optical Core and Ciena Catches Packet/Optical Convergence Bug.)

Verizon will probably pick two sources, Genovese writes, and it's possible those two will duke it out for primary status. The systems also fit Verizon's timeframe, as they'd have all the required features, including 100Gbit/s interfaces, ready for trials by late 2011.

About that timeframe -- Verizon is saying it wants to commercially deploy the boxes in 2012, but Genovese thinks that's optimistic. (It'll be hard to get anything done that year anyway, with the world ending and all.) He's thinking the real OTP rampup might not happen until 2013.

Separately, Genovese notes that Verizon's new architecture isn't as anti-core-router as some have believed. "While there have been some studies suggesting up to a 60 percent reduction in core router ports with the OTP architecture, we believe this is overstated. Based on our industry conversations we think there could be up to a 10 percent to 15 percent reduction," he writes.

— Craig Matsumoto, West Coast Editor, Light Reading

Pete Baldwin 12/5/2012 | 4:22:07 PM
re: AlcaLu, Ciena Look Good for Verizon RFP

Neglected to mention: Genovese is estimating this OTP stuff could garner vendors "hundreds of millions in revenue" over several years.

obaut 12/5/2012 | 4:22:07 PM
re: AlcaLu, Ciena Look Good for Verizon RFP

So what really is packet-optical about this FRP?

There are non-packet-aware fiber-optic transport interfaces, and there is a non-optical packet switch fabric. But these factors alone are not a reason to call this now P-OTS; othw, any single platform mechanically integrating L2 switching and WDM could have been labeled P-OTS for a number of years.

Are there actually new packet-aware optical features called for in this RFP? What are the specifics about ASON/ASTN in this regard?

Or is the main purpose of this RFP simply to get more L2-0 capacity at lower cost per bps?

Moreover, is there any intent in this FRP to react to the fact most of the current packet traffic no longer is plain data file transfer, but increasingly stream oriented and interactive communications such as video conf etc needing QoS best achieved by direct L1-0 transport? Are the P-OTS platforms to more cost-effectively support these new packet based services, other than through plain lower cost bulk capacity?

With video etc. bandwidth intensive multimedia dominating traffic growth, it looks like the service providers have an exponential capacity and economical problem to deal with, and plain linear cost/bps reduction measures through bulk capacity increases would not seem to provide a viable solution.


Sterling Perrin 12/5/2012 | 4:21:58 PM
re: AlcaLu, Ciena Look Good for Verizon RFP


Verizon has labeled this project as LH-OTP so they haven't particularly played up the packet side so far - so you can blame us (Heavy Reading/Light Reading) for some of the packet-optical hype... However, their concept is DWDM+OTN+MPLS-TP for the core so there is a packet-optical integration required. Also, OTN switching is a means to handle Sonet/SDH (TDM) and Ethernet traffic - so, in my view, this is a packet-optical trend as well. I understand that OTN is a TDM technology, but with advances such as ODUflex and ODU0, it has been adapted specfically to handle Ethernet traffic - such as that generated throught high-speed Internet, video, etc.


tmmarvel 12/5/2012 | 4:21:35 PM
re: AlcaLu, Ciena Look Good for Verizon RFP

On the topic - do the GMPLS/ODUflex specs in the RFP include Hitless Adjustment of ODUflex (ITU-T G.HAO) features?

Sign In