Dropping from 4096QAM to 2048 or 1024 is a first world problem.

 

 

From: AF <[email protected]> On Behalf Of Steve Jones
Sent: Thursday, August 27, 2020 8:33 AM
To: AnimalFarm Microwave Users Group <[email protected]>
Subject: Re: [AFMUG] Upstream Bandwidth Question

 

I have a link I'm considering like that right now. 4096 has like 27 days annual 
out but I still stay over a gig. 

It brings a ring in and still is better thanni get on my ptp650 and force 200 
link

 

On Wed, Aug 26, 2020, 4:23 PM Ken Hohhof <[email protected] 
<mailto:[email protected]> > wrote:

It depends on the frequency and duration of fiber cuts / outages.  I used to 
think reliability was everything.  Lately, customers whine about bandwidth 
constantly.  If there is a several hour outage once or twice a year, they whine 
a little and then forget it.  But speed is at the top of the hierarchy of 
needs, above things like food and sex.

 

Actually I am adapting this from my approach to licensed links.  I used to 
think every link had to be designed for 5 nines (while realizing that with 
climate change that probably got us 4 nines IRL).  So if I couldn’t find an 
intermediate point, or use 6 ft dishes, I’d do things like use a 30 MHz channel 
width to get a little incremental system gain.  Now I will go for the most 
capacity on every link, figuring more bandwidth 363 days a year was more 
important than 15 minutes of downtime 2 days a year.  And if I don’t need all 
that capacity today, I’ll need it in the near future.  The need for speed is 
not going away.

 

 

From: AF <[email protected] <mailto:[email protected]> > On Behalf 
Of [email protected] <mailto:[email protected]> 
Sent: Wednesday, August 26, 2020 4:07 PM
To: AnimalFarm Microwave Users Group <[email protected] <mailto:[email protected]> >
Subject: Re: [AFMUG] Upstream Bandwidth Question

 

I prefer multiple ingress/egress points. You could have 100 different upstreams 
but if they go across one fiber line, you are screwed.

 

On Wed, Aug 26, 2020 at 4:50 PM Mike Hammett <[email protected] 
<mailto:[email protected]> > wrote:

I'm always in favor of more diversity. I'd go with an additional wavelength, 
even if it's out of the same POP you're already in, just going somewhere else 
(and is diverse as possible, when using a single POP.

 

Bigger pipes going to the same places (assuming the small pipes are sufficient 
to handle what you need) just isn't as good of a design.

 



-----
Mike Hammett
 <http://www.ics-il.com/> Intelligent Computing Solutions
 <https://www.facebook.com/ICSIL>  
<https://plus.google.com/+IntelligentComputingSolutionsDeKalb>  
<https://www.linkedin.com/company/intelligent-computing-solutions>  
<https://twitter.com/ICSIL> 
 <http://www.midwest-ix.com/> Midwest Internet Exchange
 <https://www.facebook.com/mdwestix>  
<https://www.linkedin.com/company/midwest-internet-exchange>  
<https://twitter.com/mdwestix> 
 <http://www.thebrotherswisp.com/> The Brothers WISP
 <https://www.facebook.com/thebrotherswisp>  
<https://www.youtube.com/channel/UCXSdfxQv7SpoRQYNyLwntZg> 





  _____  


From: "Mark - Myakka Technologies" <[email protected] <mailto:[email protected]> >
To: [email protected] <mailto:[email protected]> 
Sent: Wednesday, August 26, 2020 1:32:46 PM
Subject: [AFMUG] Upstream Bandwidth Question

We  are  starting  to  get close enough on upstream bandwidth, where I
need to start thinking about getting some more.

Currently  we have two 10G ports going to  two different data centers.
With  the  internet  connections, IX links and current hardware, I can
pull  about 8GB max from each port.  Currently in peak times we pull a
total  of  about  4-5 gbps on one line and about 3-4 gbps on the other.
 It  is manageable now, until an upstream goes down during peak times.
 That has only happened once and it was a bit dicey.

 The question is which would be "better"?

 A.  Upgrade the two pipes to 20/40/100G whatever is the standard now.
 Buy new hardware to handle the new speeds needed.

 B.   Build   out   a  9  mile fiber run to a possible POP where I can
 possibly grab some dark fiber or at least another 10G and run to a possible 
3rd data center.

 I  personally  like B, being I can push upgrading the other two links
 down  the road.  This is a rural area, but a state road, so the fiber
 install  won't  be  cheap.   However,  I  can  possibly pick up 10-15
 customers along the way with the possibility of getting more as things
 build out.

 Currently  the  two upstreams  balance themselves out.  I don't
 have  any  fancy  code doing load balancing.  If I add a third to the
 mix, I'm not sure how well they will balance out.

 Just trying to figure out pros and cons of each.


--

Thanks,
 Mark                          mailto:[email protected] <mailto:[email protected]> 

Myakka Technologies, Inc.
www.Myakka.com <http://www.Myakka.com> 


-- 
AF mailing list
[email protected] <mailto:[email protected]> 
http://af.afmug.com/mailman/listinfo/af_af.afmug.com

 

-- 
AF mailing list
[email protected] <mailto:[email protected]> 
http://af.afmug.com/mailman/listinfo/af_af.afmug.com

-- 
AF mailing list
[email protected] <mailto:[email protected]> 
http://af.afmug.com/mailman/listinfo/af_af.afmug.com

-- 
AF mailing list
[email protected]
http://af.afmug.com/mailman/listinfo/af_af.afmug.com

Reply via email to