[AusNOG] 10GBase-T SFP modules?

James Andrewartha trs80 at ucc.gu.uwa.edu.au
Thu Jul 10 13:17:32 EST 2014


On Thu, 10 Jul 2014, Beeson, Ayden wrote:

> Agreed, with Wave 1 it's basically a "nice to have" at the most, in a lot of cases, fairly redundant.

I was talking about wave 2.

> Wave 2 will change that though, with theoretical speeds going much 
> higher with 160 MHz channels and beamforming to allow 3+ Gbps.

There's a single 160Mhz channel available, no enterprise is going to use 
it. There's only two 80MHz channels that aren't affected by weather radar. 
So real business deployments will use 40MHz channels to avoid channel 
overlap.

> The AP will absolutely need to deliver more than 1gbps to reach anywhere 
> near its speed allowances, even if that bandwidth is spread across 
> multiple users.

And what applications need those speed allowances?

Don't get me wrong, I think the PHY improvements in 802.11ac are great for 
increased speed, but I can hardly see the need to cater for > 1Gbps real 
througput. I don't see companies paying four times the price for 10GBASE-T 
switches just so their APs can gain marginal improvements in peak 
performance.

Recent Wave 1 APs have reduced power enough to fit into 802.3af, but I bet 
Wave 2 will require 802.3at for 4SS, which a large amount of the installed 
base of switches don't support. PoE over 10GBASE-T isn't even ratified yet 
from what I can tell, although it does work OK according to this report 
from May: http://www.ieee802.org/3/bt/public/may14/darshan_2_0514.pdf

-- 
# TRS-80              trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \
# UCC Wheel Member     http://trs80.ucc.asn.au/ #|  what squirrels do best     |
[ "There's nobody getting rich writing          ]|  -- Collect and hide your   |
[  software that I know of" -- Bill Gates, 1980 ]\  nuts." -- Acid Reflux #231 /


More information about the AusNOG mailing list