The evolution that will start the revolution

You’ve heard it all before, evolutionary technology versus revolutionary technology. Everyone wants their newest technology to be revolutionary – expecting it to be life changing and a wide-sweeping, compelling reason to spend tons of money. This is rarely the case and more often than not marketing fluff to try and get you onto the next big thing. Occasionally we get such an unassuming technology announcement that fits squarely in the ‘no big deal’ from a speeds and feeds perspective that it’s easy to overlook. This is clearly the case with the recent multigigabit announcements from Cisco during Cisco Live, Milan. Multigigabit is a technology that allows your existing cabling to support speeds in excess of 1G, without having to make the jump all the way to significantly more costly 10G. Since we already have technology that address speeds and feeds above what we’re talking about here (how many 10G uplinks have you deployed recently?), it’s easy to overlook the impact this will bring to our daily lives. The ability to move to 2.5G and 5G link speeds without having to make the jump all the way to 10G will allow us to get improved link speeds without having to pay a premium for them. The expected cost increase is estimated to be anywhere from 0% to 15% according to the rumor mill which makes the 250% to 500% speed bump quite attractive!

802.11ac wave 2
The reason I’m taking about it is the fact that this brings with it the promise of addressing the 1G bottleneck that people have been gnashing their teeth over in the wireless space for the past couple of years. While we’ve been able to reasonably deflect the speeds and feeds conversation with 802.11ac wave 1 (speeds approaching 1G wired requirements), there has been no good way to move past that without having a two-cable conversation. The assumption up till now has been that 2x 1G links will be the way forward and many people have been running two copper runs out to their Access Points for the past several years in anticipation of this approach. 802.11ac wave 2 will undoubtedly break the 1G barrier in fairly short order with speeds being promised of (best case) 6,930Mbps PHY rate (about 4,900Mbps on the wire). Multigigabit solutions will allow us to address these concerns without having to invest in 10G links. Better yet, it will allow us to address these concerns without having to consume two 1G ports on our switches. Regardless of the solution you choose (1x 10G or 2x 1G), the cost for deploying a single Multigigabit link supporting up to 5G will be less at scale.

The other unassuming byproduct of this conversation is that Access Points require power to bring up all of those components. It will be nearly impossible to power up a 10G ethernet interface in an AP in the power budgets that we have today. By reducing the link speed requirements to 5G, we can save power at the edge device and still fit in modern negotiation. Multigigabit solutions today will provide PoE, PoE+, and UPoE to ensure that the wave 2 APs that we’re going to be hanging will have ample power for whatever they’re going to bring.

The Revolution
I predict that the incremental cost and intermediary speeds will allow us to start having conversations about the jump to 10G. Multigigabit solutions on Access Points, switch uplinks, and desktop and server nics will be the next big thing. Stackable solutions today promise backwards compatibility so you don’t have to rip and replace – just add a stack member and you’re good to go in that closet/IDF! Regardless of your future proofing plans, enabling faster wireless, or just ensuring that you’re not spending money after (can you believe it?) now legacy 1G infrastructure, make sure you’re having a conversation today about ethernet to bridge the gap to 10G.

For more information about the NBASE-T alliance, go here.
For the Cisco Live, Milan – Tech Field Day Extra event with Peter Jones, go here.

10 Responses to The evolution that will start the revolution

  1. wirednot says:

    Reblogged this on wirednot and commented:
    I like Sam’s take on something I demanded early on in the 11ac arc. Have a read, see if you don’t agree.

  2. Pingback: The evolution that will start the revolution - Tech Field Day

  3. Nolan says:

    I can has 5G/10G internets too pl0x?


    • scwifi says:

      Yes Nolan, you certainly can. While I assume your comment was mostly tongue-in-cheek, the point is that access layers will always outstrip our internet connections. Moore’s law mandates that we get bigger, better, faster for cheaper on a regular basis. While it may not manifest itself in WAN links all the times, the local speeds for file transfers and streaming media on the LAN is still a viable need for speeds beyond 1G.

  4. Pingback: 3/4/2015: Sticky Clients, New Tools, Help Manage the Message (and the Environment) | W to the 4th

  5. Nice write up Sam. I’m starting to use the work “pragmatic” a lot these days. I also like the evolution vs revolution distinction.
    As an FYI – I’m head to Berlin next week for the IEEE 802.3 meeting where we hope to move from “study group” (talking about what we should do) to “task force” (doing it) which will let us get really started on getting an IEEE 802.3 standard written – see to the current Study group public docs.

    • scwifi says:

      Thanks Peter! I appreciate the comments and look forward to keeping tabs on the formal standard as it evolves and comes to fruition!

  6. Jon foster says:

    In reality i doubt we’ll see those speeds for some time. Enterprise Wi-Fi is very unlikely to use the extra wide channels required and i dont see any monster 8 stream APs or clients anywhere yet (ever). 802.11n’s theoretical 4 streams also never appeared…

    So IMO, no great rush for greater than 1 gig for wifi

    • scwifi says:

      Fair point Jon! I think there will always be an oversubscription component to design conversations but there will also be those that don’t want to talk about it, they just want to make sure the pipes are as logically sane for the long term as much as possible. Many customers expect in excess of 5 years out of their infrastructure today and I think it’s safe to say that we have no clue what x link will look like in that period of time – much less 7 or even 10 years out. Regardless, you bring up a good issue which is the fact that the designs we implement today are going to be with us for a long time. Let’s make sure we collectively set our expectations long term – even if we don’t know what they’re going to be. 🙂

  7. Pingback: Cisco Live Europe: NBASE-T progress report | Just another day at the office...

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: