Act Two, wherein OPTICAL TRANSPORT (OT) discusses with Ethernet Services (ES) how it can take advantage of Flexible Ethernet interfaces to provide a more efficient end-to-end transport, thus describing what is termed FlexE Aware Transport and FlexE Terminated Transport.
ES: Last time we talked, I told you all about my new FlexE capabilities, and described three use cases, Bonding, Channelization, and Sub-rating.
OT: And you pointed out that you could apply these use cases over my optical transport facilities, without my being aware that you were doing so.
ES: Correct. My FlexE interfaces would appear to you as ordinary Ethernet traffic.
OT: We decided to call that FlexE Unaware Transport.
ES: Your assignment was to investigate what you might do differently if you were “aware” you were transporting FlexE traffic.
OT: I have done my homework. It seems that I can mostly easily add value in the Sub-rating use case.
ES: That is where I am giving you only a portion of the traffic the physical interface can handle.
OT: Exactly, but in an un-aware situation, you are paying for the full physical capacity. It would benefit both of us if I only needed to transport the capacity you were using. I would not waste bandwidth, and you would only need to pay for what you use.
ES: Sounds good to me. Tell me more.
OT: Let’s look at the example we discussed last time. You needed to transport a 150GE interface, but you didn’t have a physical interface to support that, and even if you did, I didn’t have a client interface for that.
ES: Correct. I transmitted the interface over two 100G interfaces, each one sub-rated to 75G. As you said, I needed to pay for 2 x 100G of transport bandwidth.
OT: Well if I am “aware” that you are sub-rating the traffic, then I can handle it more efficiently. But I need a mechanism to tell me what portion of the bits you feed me are valid and what are junk?
ES: I can do that. My FlexE Shim has a “calendaring” mechanism, where I arrange the bits in a series of time slots. In a sub-rating case, I fill up certain slots with the bits I want transported, and then fill the others with null bits. I have no problem making you aware in our interface what part of the calendar has valid bits.
OT: That sounds workable. I can implement a “Partial FlexE Shim” in my transponders and muxponders. This would communicate with your calendar, and then I would only need to transport the good bits, and ignore the null bits. I would call this FlexE Aware Transport.
ES: I like this approach. I take back what I said when we last spoke. Is there anything else you can do for me?
OT: In fact there is. I can actually terminate your FlexE Shim in my network. I would call this FlexE Terminated Transport.
ES: How would this help me?
OT: This can make applications of Channelization much more efficient, by using my OTN switching capabilities.
ES: Can you give me an example.
OT: Sure. Say you have three routers that you want to connect in a triangular fashion using non-standard transport rates. Today you need to connect each route individually, and have a lot of wasted capacity.
ES: Well that’s because I am using transport in a transparent fashion.
OT: Exactly, but by terminating your shims with my own, and coupling this with my flexible transport capabilities I make life easier for you on two fronts. First, I reduce the number of physical connections you to make to the transport network. Second, similar to the sub-rating discussion example above, I make sure you only consume, and pay for, the transport capacity required for each route.
ES: It certainly does look like when we communicate that we can work together to a win-win.
To learn more about ECI’s 5G Solutions, click here.
Jonathan Homa, is Senior Director Portfolio Marketing at ECI for the last few years. Jonathan enjoys French beer in Nice and loves to discuss the 3 hottest topics in networking today: softwarization, virtualization and cloudification.