From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from vsmx002.dclux.xion.oxcs.net (vsmx002.dclux.xion.oxcs.net [185.74.65.108]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 3BFF43CB37; Fri, 17 Mar 2023 14:22:19 -0400 (EDT) Received: from proxy-2.proxy.oxio.ns.xion.oxcs.net (proxy-2.proxy.oxio.ns.xion.oxcs.net [83.61.18.4]) by mx-out.dclux.xion.oxcs.net (Postfix) with ESMTPA id 51AF28C0D5F; Fri, 17 Mar 2023 18:22:12 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=dclux.xion.oxcs.net; s=mail1; t=1679077338; bh=iHYG72ms5O4XeWmwkOQYPREpMbrGIvNTzrd6eJUZxTo=; h=Date:From:To:Cc:In-Reply-To:References:Subject:From; b=Vla2GUhvq0qzthiTCzvP//cOKGrBJqe4ma03QF2S4296E88LifSoYfbLGGdlmQ3tN WOEfPT/nhR4yIoSGtPUiMw1o5Dmpd6ZqZVx3lU2c/hEKkyMMUi5sXB/Pe8TFSPGE6R jMR72RAIGUXG8JQhh8XmL3d+rZ4QZNa3FLmMT0hr7irC3mdGOp3lB5NcuepmU06cLb Czgn/+yvVcBFfXZaXVVbb/nvA7zAlmTogMft87TVfEuXR8TwGJWPYW58bw19N6q1qn Hk+R8g/mte2SioLtDCaWIES4aftpBJiRWCsK5KyCsWI58Lq3OLsljgu1wrmNpafu8c PhmRf4ekUB9rw== Date: Fri, 17 Mar 2023 19:21:53 +0100 From: Mike Puchol To: Dave Taht Cc: Dave Taht via Starlink , Rpm , libreqos , bloat Message-ID: In-Reply-To: References: <1672786712.106922180@apps.rackspace.com> <77CCAD19-07E0-4F9E-88C1-D207CF7BF376@cable.comcast.com> <83ffc0dad19e3343e49271889369cefc@rjmcmahon.com> <3CD0B9E6-0B2A-4A70-8F53-ED0822DF77A6@gmx.de> <13DE6E53-665F-4C20-BBE2-70E685421E9D@gmx.de> <22C819FA-DDD7-4B9B-8C09-8008D4273287@gmx.de> <5e7fac51071bdbb20837e72e7eedfc7c@rjmcmahon.com> <3f45d2a0b6e46d7b2775fb801e805f93@rjmcmahon.com> <70F71290-C6CB-4D19-8A88-F0F17C0BDDA2@gmx.de> <5e0cd693c4749d128dbb48d6c1129071@rjmcmahon.com> <2ab2983d-6beb-49cb-8c35-e481cbfdc7a3@Spark> X-Readdle-Message-ID: e5047219-c41a-4c40-a293-7a75e623469d@Spark MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="6414afd2_4db127f8_1631" X-VadeSecure-Status: LEGIT X-VADE-STATUS: LEGIT Subject: Re: [Bloat] [Rpm] [Starlink] On FiWi X-BeenThere: bloat@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: General list for discussing Bufferbloat List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Mar 2023 18:22:19 -0000 --6414afd2_4db127f8_1631 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline A four-port EPON OLT with modules goes for =24500, serves up to 256 custo= mers. To serve same amount you need 36 netPowers, at =24140 each, total CAPEX =24= 5,000. What you then spend on PON splitters you also spend on PoE injectors for = the netPower, and drop cable is cheaper than Ethernet (at least if you wa= nt it to send power further than 10 meters=E2=80=A6 no CCA allowed). It=E2=80=99s not so clear-cut, each can fit a certain deployment scenario= , so I would never argue in antagonistic terms. Best, Mike On Mar 17, 2023 at 17:38 +0100, Dave Taht , wrote:= > This is a pretty neat box: > > https://mikrotik.com/product/netpower=5Flite=5F7r > > What=C2=A0are the compelling arguments for fiber vs copper, again=3F > > > > On Tue, Mar 14, 2023 at 4:10=E2=80=AFAM Mike Puchol via Rpm wrote: > > > Hi Bob, > > > > > > You hit on a set of very valid points, which I'll complement with m= y views on where the industry (the bit of it that affects WISPs) is headi= ng, and what I saw at the MWC in Barcelona. Love the =46iWi term :-) > > > > > > I have seen the vendors that supply WISPs, such as Ubiquiti, Cambiu= m, and Mimosa, but also newer entrants such as Tarana, increase the perfo= rmance and on-paper specs of their equipment. My examples below are cente= red on the African market, if you operate in Europe or the US, where you = can charge customers a higher install fee, or even charge them a break-up= fee if they don't return equipment, the economics work. > > > > > > Where currently a =7E=24500 sector radio could serve =7E60 endpoint= s, at a cost of =7E=2450 per endpoint (I use this term in place of ODU/CP= E, the antenna that you mount on the roof), and supply =7E2.5 Mbps CIR pe= r endpoint, the evolution is now a =7E=242,000+ sector radio, a =24200 en= dpoint, capability for =7E150 endpoints per sector, and =7E25 Mbps CIR pe= r endpoint. > > > > > > If every customer a WISP installs represents, say, =24100 CAPEX at = install time (=2450 for the antenna + cabling, router, etc), and you char= ge a =2430 install fee, you have =2470 to recover, and you recover from t= he monthly contribution the customer makes. If the contribution after OPE= X is, say, =2410, it takes you 7 months to recover the full install cost.= Not bad, doable even in low-income markets. > > > > > > =46ast-forward to the next-generation version. Now, the CAPEX at in= stall is =24250, you need to recover =24220, and it will take you 22 mont= hs, which is above the usual 18 months that investors look for. > > > > > > The focus, thereby, has to be the lever that has the largest effect= on the unit economics - which is the per-customer cost. I have drawn wha= t my ideal =46iWi network would look like: > > > > > > > > > > > > Taking you through this - we start with a 1-port, low-cost EPON OLT= (or you could go for 2, 4, 8 ports as you add capacity). This OLT has ca= pacity for 64 ONUs on its single port. Instead of connecting the typical = fiber infrastructure with kilometers of cables which break, require maint= enance, etc. we insert an EPON to Ethernet converter (I added =22magic=22= because these don't exist A=46AIK). > > > > > > This converter allows us to connect our =242k sector radio, and ser= ve the =24200 endpoints (ODUs) over wireless point-to-multipoint up to 10= km away. Each ODU then has a reverse converter, which gives us EPON again= . > > > > > > Once we are back on EPON, we can insert splitters, for example, pre= -connectorized outdoor 1:16 boxes. Every customer install now involves a = 100 meter roll of pre-connectorized 2-core drop cable, and a =2420 EPON O= NU. > > > > > > Using this deployment method, we could connect up to 16 customers t= o a single =24200 endpoint, so the enpoint CAPEX per customer is now =241= 2.5. Add the ONU, cable, etc. and we have a per-install CAPEX of =2482.5 = (assuming the same =2450 of extras we had before), and an even shorter br= eak-even. In addition, as the endpoints support higher capacity, we can p= rovision at least the same, if not more, capacity per customer. > > > > > > Other advantages: the =24200 ODU is no longer customer equipment an= d CAPEX, but network equipment, and as such, can operate under a longer b= reak-even timeline, and be financed by infrastructure PE funds, for examp= le. As a result, churn has a much lower financial impact on the operator.= > > > > > > The main reason why this wouldn't work today is that EPON, as we kn= ow, is synchronous, and requires the OLT to orchestrate the amount of tim= e each ONU can transmit, and when. Having wireless hops and media convers= ions will introduce latencies which can break down the communications (e.= g. one ONU may transmit, get delayed on the radio link, and end up overla= pping another ONU that transmitted on the next slot). Thus, either the =22= magic=22 box needs to account for this, or an new hybrid EPON-wireless pr= otocol developed. > > > > > > My main point here: the industry is moving away from the unconnecte= d. All the claims I heard and saw at MWC about =22connecting the unconnec= ted=22 had zero resonance with the financial drivers that the unconnected= really operate under, on top of IT literacy, digital skills, devices, po= wer... > > > > > > Best, > > > > > > Mike > > > On Mar 14, 2023 at 05:27 +0100, rjmcmahon via Starlink , wrote: > > > > To change the topic - curious to thoughts on =46iWi. > > > > > > > > Imagine a world with no copper cable called =46iWi (=46iber,VCSEL= /CMOS > > > > Radios, Antennas) and which is point to point inside a building > > > > connected to virtualized APs fiber hops away. Each remote radio h= ead > > > > (RRH) would consume 5W or less and only when active. No need for = things > > > > like zigbee, or meshes, or threads as each radio has a fiber conn= ection > > > > via Corning's actifi or equivalent. Eliminate the AP/Client power= > > > > imbalance. Plastics also can house smoke or other sensors. > > > > > > > > Some reminders from Paul Baran in 1994 (and from David Reed) > > > > > > > > o) Shorter range rf transceivers connected to fiber could produce= a > > > > significant improvement - - tremendous improvement, really. > > > > o) a mixture of terrestrial links plus shorter range radio links = has the > > > > effect of increasing by orders and orders of magnitude the amount= of > > > > frequency spectrum that can be made available. > > > > o) By authorizing high power to support a few users to reach slig= htly > > > > longer distances we deprive ourselves of the opportunity to serve= the > > > > many. > > > > o) Communications systems can be built with 10dB ratio > > > > o) Digital transmission when properly done allows a small signal = to > > > > noise ratio to be used successfully to retrieve an error free sig= nal. > > > > o) And, never forget, any transmission capacity not used is waste= d > > > > forever, like water over the dam. Not using such techniques repre= sent > > > > lost opportunity. > > > > > > > > And on waveguides: > > > > > > > > o) =22=46iber transmission loss is =7E0.5dB/km for single mode fi= ber, > > > > independent of modulation=22 > > > > o) =E2=80=9CCopper cables and PCB traces are very frequency depen= dent. At > > > > 100Gb/s, the loss is in dB/inch.=22 > > > > o) =22=46ree space: the power density of the radio waves decrease= s with the > > > > square of distance from the transmitting antenna due to spreading= of the > > > > electromagnetic energy in space according to the inverse square l= aw=22 > > > > > > > > The sunk costs & long-lived parts of =46iWi are the fiber and the= CPE > > > > plastics & antennas, as CMOS radios+ & fiber/laser, e.g. VCSEL co= uld be > > > > pluggable, allowing for field upgrades. Just like swapping out S=46= P in a > > > > data center. > > > > > > > > This approach basically drives out Wi=46i latency by eliminating = shared > > > > queues and increases capacity by orders of magnitude by leveragin= g 10dB > > > > in the spatial dimension, all of which is achieved by a physical = design. > > > > Just place enough RRHs as needed (similar to a pop up sprinkler i= n an > > > > irrigation system.) > > > > > > > > Start and build this for an MDU and the value of the building imp= roves. > > > > Sadly, there seems no way to capture that value other than over l= ong > > > > term use. It doesn't matter whether the leader of the HOA tries t= o > > > > capture the value or if a last mile provider tries. The value rem= ains > > > > sunk or hidden with nothing on the asset side of the balance shee= t. > > > > We've got a CAPEX spend that has to be made up via =22OPEX return= s=22 over > > > > years. > > > > > > > > But the asset is there. > > > > > > > > How do we do this=3F > > > > > > > > Bob > > > > =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F= =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F= > > > > Starlink mailing list > > > > Starlink=40lists.bufferbloat.net > > > > https://lists.bufferbloat.net/listinfo/starlink > > > =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F= =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F > > > Rpm mailing list > > > Rpm=40lists.bufferbloat.net > > > https://lists.bufferbloat.net/listinfo/rpm > > > -- > Come Heckle Mar 6-9 at: https://www.understandinglatency.com/ > Dave T=C3=A4ht CEO, TekLibre, LLC --6414afd2_4db127f8_1631 Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline
A four-port EPON OLT with modules goes for =24500, = serves up to 256 customers.&=23160;

To serve same amount you need 36 netPowers, at =24140 each, total CAPEX =24= 5,000.&=23160;

What you then spend on PON splitters you also spend on PoE injectors for = the netPower, and drop cable is cheaper than Ethernet (at least if you wa= nt it to send power further than 10 meters=E2=80=A6 no CCA allowed).

It=E2=80=99s not so clear-cut, each can fit a certain deployment scenario= , so I would never argue in antagonistic terms.

Best,

Mike
On Mar 17, 2023 at 17:38 +0100, Dav= e Taht <dave.taht=40gmail.com>, wrote:
This is a pretty neat box:


What&=23160;are the compelling arguments for fiber vs copper, again=3F=


On Tue, Mar 14, 2023 at 4= :10=E2=80=AFAM Mike Puchol via Rpm <rpm=40lists.bufferbloat.net> wrote:
Hi Bob,

You hit on a set of very valid points, which I'll complement with my view= s on where the industry (the bit of it that affects WISPs) is heading, an= d what I saw at the MWC in Barcelona. Love the =46iWi term :-)

I have seen the vendors that supply WISPs, such as Ubiquiti, Cambium, and= Mimosa, but also newer entrants such as Tarana, increase the performance= and on-paper specs of their equipment. My examples below are centered on= the African market, if you operate in Europe or the US, where you can ch= arge customers a higher install fee, or even charge them a break-up fee i= f they don't return equipment, the economics work.

Where currently a =7E=24500 sector radio could serve =7E60 endpoints, at = a cost of =7E=2450 per endpoint (I use this term in place of ODU/CPE, the= antenna that you mount on the roof), and supply =7E2.5 Mbps CIR per endp= oint, the evolution is now a =7E=242,000+ sector radio, a =24200 endpoint= , capability for =7E150 endpoints per sector, and =7E25 Mbps CIR per endp= oint.

If every customer a WISP installs represents, say, =24100 CAPEX at instal= l time (=2450 for the antenna + cabling, router, etc), and you charge a =24= 30 install fee, you have =2470 to recover, and you recover from the month= ly contribution the customer makes. If the contribution after OPEX is, sa= y, =2410, it takes you 7 months to recover the full install cost. Not bad= , doable even in low-income markets.

=46ast-forward to the next-generation version. Now, the CAPEX at install = is =24250, you need to recover =24220, and it will take you 22 months, wh= ich is above the usual 18 months that investors look for.

The focus, thereby, has to be the lever that has the largest effect on th= e unit economics - which is the per-customer cost. I have drawn what my i= deal =46iWi network would look like:


<Hybrid EPON-Wireless network.png>
Taking you through this - we start with a 1-port, low-cost EPON OLT (or y= ou could go for 2, 4, 8 ports as you add capacity). This OLT has capacity= for 64 ONUs on its single port. Instead of connecting the typical fiber = infrastructure with kilometers of cables which break, require maintenance= , etc. we insert an EPON to Ethernet converter (I added =22magic=22 becau= se these don't exist A=46AIK).

This converter allows us to connect our =242k sector radio, and serve the= =24200 endpoints (ODUs) over wireless point-to-multipoint up to 10km awa= y. Each ODU then has a reverse converter, which gives us EPON again.

Once we are back on EPON, we can insert splitters, for example, pre-conne= ctorized outdoor 1:16 boxes. Every customer install now involves a 100 me= ter roll of pre-connectorized 2-core drop cable, and a =2420 EPON ONU.&=23= 160;

Using this deployment method, we could connect up to 16 customers to a si= ngle =24200 endpoint, so the enpoint CAPEX per customer is now =2412.5. A= dd the ONU, cable, etc. and we have a per-install CAPEX of =2482.5 (assum= ing the same =2450 of extras we had before), and an even shorter break-ev= en. In addition, as the endpoints support higher capacity, we can provisi= on at least the same, if not more, capacity per customer.

Other advantages: the =24200 ODU is no longer customer equipment and CAPE= X, but network equipment, and as such, can operate under a longer break-e= ven timeline, and be financed by infrastructure PE funds, for example. As= a result, churn has a much lower financial impact on the operator.
=
The main reason why this wouldn't work today is that EPON, as we know, is= synchronous, and requires the OLT to orchestrate the amount of time each= ONU can transmit, and when. Having wireless hops and media conversions w= ill introduce latencies which can break down the communications (e.g. one= ONU may transmit, get delayed on the radio link, and end up overlapping = another ONU that transmitted on the next slot). Thus, either the =22magic= =22 box needs to account for this, or an new hybrid EPON-wireless protoco= l developed.

My main point here: the industry is moving away from the unconnected. All= the claims I heard and saw at MWC about =22connecting the unconnected=22= had zero resonance with the financial drivers that the unconnected reall= y operate under, on top of IT literacy, digital skills, devices, power...=

Best,

Mike
On Mar 14, 2023 at 05:27 +0100, rjm= cmahon via Starlink <starlink=40lists.bufferbloat.net>, = wrote:
To change the topic - curious to thoughts on = =46iWi.

Imagine a world with no copper cable called =46iWi (=46iber,VCSEL/CMOS Radios, Antennas) and which is point to point inside a building
connected to virtualized APs fiber hops away. Each remote radio head
(RRH) would consume 5W or less and only when active. No need for things like zigbee, or meshes, or threads as each radio has a fiber connection via Corning's actifi or equivalent. Eliminate the AP/Client power
imbalance. Plastics also can house smoke or other sensors.

Some reminders from Paul Baran in 1994 (and from David Reed)

o) Shorter range rf transceivers connected to fiber could produce a
= significant improvement - - tremendous improvement, really.
o) a mixture of terrestrial links plus shorter range radio links has the<= br /> effect of increasing by orders and orders of magnitude the amount of
frequency spectrum that can be made available.
o) By authorizing high power to support a few users to reach slightly
longer distances we deprive ourselves of the opportunity to serve the
many.
o) Communications systems can be built with 10dB ratio
o) Digital transmission when properly done allows a small signal to
= noise ratio to be used successfully to retrieve an error free signal.
o) And, never forget, any transmission capacity not used is wasted
forever, like water over the dam. Not using such techniques represent
lost opportunity.

And on waveguides:

o) =22=46iber transmission loss is =7E0.5dB/km for single mode fiber,
independent of modulation=22
o) =E2=80=9CCopper cables and PCB traces are very frequency dependent. At=
100Gb/s, the loss is in dB/inch.=22
o) =22=46ree space: the power density of the radio waves decreases with t= he
square of distance from the transmitting antenna due to spreading of the<= br /> electromagnetic energy in space according to the inverse square law=22
The sunk costs & long-lived parts of =46iWi are the fiber and the CPE=
plastics & antennas, as CMOS radios+ & fiber/laser, e.g. VCSEL co= uld be
pluggable, allowing for field upgrades. Just like swapping out S=46P in a=
data center.

This approach basically drives out Wi=46i latency by eliminating shared queues and increases capacity by orders of magnitude by leveraging 10dB in the spatial dimension, all of which is achieved by a physical design.<= br /> Just place enough RRHs as needed (similar to a pop up sprinkler in an
irrigation system.)

Start and build this for an MDU and the value of the building improves. Sadly, there seems no way to capture that value other than over long
term use. It doesn't matter whether the leader of the HOA tries to
capture the value or if a last mile provider tries. The value remains
sunk or hidden with nothing on the asset side of the balance sheet.
= We've got a CAPEX spend that has to be made up via =22OPEX returns=22 ove= r
years.

But the asset is there.

How do we do this=3F

Bob
=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F= =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F
Starlink mailing list
Starlink=40lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/starlink
<= /blockquote>
=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F= =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F
Rpm mailing list
Rpm=40lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/r= pm


--
Come Heckle Mar 6-9 at: https://www.understandinglatency.co= m/&=23160;
Dave T=C3=A4ht CEO, TekLibre, LLC
--6414afd2_4db127f8_1631--