Facebook Tries Automating the Optical Network

Facebook has some ideas for automating the management of the optical network, but it's a trickier task than automating packet networks, because of the differences between vendors' implmentations.

Still, it's doable, and the key is the mixture of an abstraction layer and a lot of Facebook programming knowhow, said Sri Bala, a Facebook principal architect. He presented the company's ideas, most of which are still in the lab, last week during the Next Generation Optical Networking conference, held in Burlingame, Calif.

Some of this work could be taken open source, but so far, Facebook is doing it all solo. "We have a lot of internal tools, which gives us sufficient compute power," Bala said -- quite the understatement, given that Facebook has enough compute power to run, you know, Facebook.

The idea behind Facebook's automation work was to get engineers away from maintaining the optical network. "At Facebook, we believe in engineers building robots. Robots manage then network," Bala said.

The key is abstraction -- masking the differences between vendors' gear by using a common abstraction layer to talk to all of them. Whoever or whatever is making decisions, be it a human operator or, say, an SDN controller, would talk only to the abstraction layer, while software in that layer would send translated commands to the network management systems (NMS) in various pieces of gear.

Unlike packet networks, optical networks have to take into consideration things like wavelength assignments and optical power levels. Vendors handle these issues differently, so it's been easiest to use one vendor's gear for both sides of an optical link. Any changes to the network -- even at Facebook -- are still highly manual, Bala said.

Want to know more about fronthaul and backhaul options for wireless networks, especially 5G? Join us in New York City on November 10 for Light Reading's '5G Transport & Networking Strategies' event. There is still time to register.

So, Facebook is developing an NMS Adaptive Layer, which would provide that abstraction layer to talk to these different vendors' NMSs, taking advantage of the increasing use of APIs in network management.

Bala also sketched out ideas for automating some elements of decision-making. Facebook has written a Spectrum Assignment Engine, for example, which would make wavelength assignments while taking forecasted demand into account. That information would be fed through an orchestrator down to the NMS Adaptive Layer.

"The time from demands to bits is going to be much faster" that way, he said.

To truly make this work, Facebook would also have to gather telemetry from optical networking gear, using the information to make decisions about how to optimize the network. For example, deciding if, during a failover situation, an optical connection could swap to a route that has a different modulation scheme, just temporarily.

— Craig Matsumoto, Editor-in-Chief, Light Reading

kq4ym 11/13/2017 | 12:39:29 PM
Re: INFN With Facebook having in this operation to pay attention to "wavelength assignments and optical power levels," it will be very interesting to see if they can make it work well with all the various vendors and separate specs involved. Whether they'll open source this is a good question.
tojofay 11/7/2017 | 12:16:57 PM
CIMI corp public blog "The media could help here too.  Light Reading has done a number of NFV articles, including the one that I opened with.  It would be helpful if they'd cover the real issues here, including the fact that no international standards group or body with the same biases as the NFV ISG has a better chance of getting things right.  This is a software problem that software architectures and architects have to solve for us."
brooks7 11/2/2017 | 11:48:20 AM
Re: INFN  


I have a question for you that is tangentally related to these last two posts here.  As you recall, OSMINE was hated.  I mean HATED.  Imagine people talking about it in the age of Twitter.  But I have yet to hear objections to these processes from customers.  I am wondering if that is because:

1 - It seems to be free to included vendors.

2 - It seems a lot less heavy handed than the old OSMINE process.

3 - Sales can proceed without the completion of the process.

4 - Maturity of prespective by the vendors.

5 - All or a combination of any of them?

6 - Something else?

It seems to me that this provides a worse vendor lockout than OSMINE ever did.  But with so many fewer vendors, maybe it doesn't matter?






Carol Wilson 11/2/2017 | 4:59:38 AM
Re: INFN I totally agree with you that Facebook is solving for a much smaller set of applications. But the key word there is "solve." 

It's not my intention in any way to detract from Bellcore work - or that of industry forums - to address the vendor interoperability issues. The fact that the telecom industry continued making the same mistake over and over - developing standards with wiggle room that vendors would then exploit to create "features" - even as its major operators consistently asked for vendor-agnostic technology can't be easily set aside. It's a hallmark of industry history. 

Facebook comes at this from a practical viewpoint that doesn't repeat that history. 

Keebler 11/1/2017 | 3:41:43 PM
Re: INFN If I had worked for Telcordia, I might be offended by the suggestion that it took Facebook to figure out how to make the optical layer interoperable. After all, that's what all of those OSMINE OSSs were designed to do for many, many years.

Then, as today, the end user defined the application that they wanted to use and the software was written to adapt the unique capabilities of the equipment to the needs of the end user. Often that required interfacing to an NMS or EMS. And, as some of us old folks may recall, that process resulted in networks that were rarely if ever at the cutting edge of technology. The software could not possibly keep up with the innovations that were (and are) happening in optical technology.

Telcordia wasn't the only one, of course, just the largest and most reviled. There were numerous in-house and third-party OSS providers who build such systems, either for simple alarm consolidation or any of a dozen other more advanced applications.

Facebook is not different, just currently sexier. And they have defined a much, much smaller set of applications - point to point bandwidth transfer - that make this an easier problem to solve. And software in general is more advanced and modular, which should allow innovations to flow through more quickly.

There is very little truly new in optics interoperability management, but we are getting better at it over time. All of us. Not just Facebook.
Carol Wilson 11/1/2017 | 4:14:37 AM
Re: INFN The lack of interoperability in the optical network space has plagued telecom operators for years - interesting that Facebook might develop the solution to this issue. I wonder why that has to be?
danielcawrey 10/31/2017 | 3:04:18 PM
Re: INFN Would be interesting to learn more about this when Facebook open sources it. I would guess that's what they woulld do at some point. It's similar to what they did with datacenter tech - they opened that up to the world to positive feedback. 
Sign In