I agree with you 100% that different usages will have different uptime requirements. I'm thinking of 80 GHz from the logical perspective as a replacement for a singlemode fiber patch cable between two routers' SFP or SFP+ interfaces speaking BGP and OSPF with each other.
In the Seattle area we have done up to 2.5 km at five nines 60cm dishes and high powered 80 GHz radios that were specifically configured to *not* reduce data rates below 1 Gbps. In the case of a severe rain fade event that would take out the link, the routers did the job of moving traffic onto backup paths or moving traffic the other direction around a ring. You can definitely go a for longer distances if your usage scenario is designed to tolerate things like dropping from 1Gbps to 100Mbps, or failing over onto a parallel backup 5 GHz link, or something like that. For ISP usage with TCP/IP and traffic flows this is inextricably linked to the layer 3 IP routing and network traffic management design - if you have a link that is constantly pushing 400-550Mbps in a sine wave pattern across a 80 GHz backbone link, and it fails over onto a 150Mbps backup parallel 5 GHz link, you're going to have a bad time. If that same link's average peak hour data flows will fit in the 150 Mbps backup link, it's a totally different situation. It's necessary to look at simultaneous occurrences of two possible events: If a major rain fade event happens during peak evening hours netflix watching time, what happens to the traffic? As Daniel said, very much dependent on usage scenario. On Wed, Oct 5, 2016 at 6:20 PM, Daniel White <afmu...@gmail.com> wrote: > Problem is your only thinking in the context of five nines wireless. > > > > No one with any common sense thinks “five nines wireless” and long range > millimeter wave. Period. I agree with you 100%. But… > > > > For instance – I engineered a BridgeWave link at 10 miles for the US Air > Force at White Sand Missile Range that was for a remote area that needed > 1Gbps (no microwave solutions at that time could do it). Since they were > playing with lasers, they told me they would never be onsite if there was > so much as a cloud in the sky. To them, it worked “five nines”. > > > > Travis Johnson also had an 18GHz link at 35 miles that worked great for > him for a number of years in his WISP. That would be against the book > too. Proper engineering and understanding the mission requirements are > critical to great engineering. > > > > Daniel White > > Managing Director – Hardware Distribution Sales > > ConVergence Technologies > > Cell: +1 (303) 746-3590 > > dwh...@converge-tech.com > > > > *From:* Af [mailto:af-boun...@afmug.com] *On Behalf Of *Eric Kuhnke > *Sent:* Wednesday, October 5, 2016 12:14 PM > *To:* af@afmug.com > *Subject:* Re: [AFMUG] Siklu Distance > > > > ha ha, maybe in ITU rain zone A (arabian desert) and with 60cm dishes, and > allowing for "uptime" in nines to include QPSK modulation at very low data > rates during rain events... > > There is a big difference between marketing materials and production > reality with five nines 80 GHz. > > > > On Wed, Oct 5, 2016 at 10:22 AM, Jason McKemie < > j.mcke...@veloxinetbroadband.com> wrote: > > I received an email from Siklu saying their equipment could offer gigabit > speeds at 8+ miles. This seems dishonest at best. I guess they didn't say > where the link had to be located to get this distance... > > > > > <https://www.avast.com/en-us/lp-safe-emailing-2270-c?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=oa-2270-c> > Virus-free. > www.avast.com > <https://www.avast.com/en-us/lp-safe-emailing-2270-c?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=oa-2270-c> >
