Re: Dedicated Server and IP anycast provider recommendation

[Étienne via NANOG]

On 07/08/18 14:49, John Kristoff wrote:

Friends,

For those that may have used or know of a service like this.  I know
some exist, but it doesn't seem to be that popular or widely advertised
as a standard service.

I'm interested in pointers to a hosting/network provider that leases
dedicated servers and can provide an anycast IP address assignment to
two or more US-diversely connected POPs, but with reasonably consistent
routing (e.g. peering, transit).  A customer-shared prefix is OK. I'm
interested in pointers to networks that would provide the prefix and
handle all the routing.
Not sure you're still looking for something, but there's this 
spreadsheet that has a few pointers: http://bgp.services/


Cheers,

--
Étienne

Re: Dedicated Server and IP anycast provider recommendation

[Damian Menscher via NANOG]

Not quite a dedicated server, but may meet your needs anyway:
https://cloud.google.com/load-balancing/

Damian

On Tue, Aug 7, 2018 at 6:50 AM John Kristoff  wrote:

> Friends,
>
> For those that may have used or know of a service like this.  I know
> some exist, but it doesn't seem to be that popular or widely advertised
> as a standard service.
>
> I'm interested in pointers to a hosting/network provider that leases
> dedicated servers and can provide an anycast IP address assignment to
> two or more US-diversely connected POPs, but with reasonably consistent
> routing (e.g. peering, transit).  A customer-shared prefix is OK. I'm
> interested in pointers to networks that would provide the prefix and
> handle all the routing.
>
> If you represent a network and sales is part of your job, I don't mind
> an off list pointer to a web page describing such a service, but please,
> this is not an invitation for "call me to discuss needs and options"
> replies nor an opportunity to get me on your customer prospect list.
> You likely ensure I never do business with you if you do either of
> those things.  :-)
>
> Thank you,
>
> John
>

Re: Dedicated Server and IP anycast provider recommendation

[John Kristoff]

On Mon, 13 Aug 2018 12:31:44 +
Étienne via NANOG  wrote:

> Not sure you're still looking for something, but there's this 
> spreadsheet that has a few pointers: http://bgp.services/

Thanks again.  This is at least the third time someone has pointed this
web page out to me.  :-)

To summarize, NetActuate and Packet were two of the most commonly
suggested providers and probably the closest to what I was asking for.

Some well known providers can do this, but don't have a specific web
page / service offering describing it as a standard product offering.

Many others, such as the often mentioned Vultr, require you to bring
your own BGP speaker and prefix.

What I was asking for isn't a popular, mass-marketed product offering,
but is available.  It just requires some research and careful
evaluation.

I think I"m familiar many hosting providers, especially the so-called
low-end/cost providers for an entirely separate project. I'm always
interested in those too, but not specifically for their ability to
speak BGP.  I maintain my own list here of providers I use here:

  

Follow ups about that list are best sent to me directly or at that page
directly.

Thanks for all the on/off list replies and suggestions.

John

Re: optical circulator as a bidirectional one fiber solution

[Eric Kuhnke]

Something that is broadly the same as a coherent 100G QPSK single
wavelength optical module, but in two different frequencies, and a passive
CWDM mux/demux prism at each end might work. The limitation would be
availability of optics for a modern 100G MSA that are both coherent and
Tx/Rx at two different THz frequencies.

Or with some box and vendor equipment in between, such as:

http://cdn.extranet.coriant.com/resources/Application-Notes/AN_Groove_Bidirectional_Fiber_74C0169.pdf?mtime=20180206023321

On Tue, Aug 7, 2018 at 1:00 PM Daniel Corbe  wrote:

> On 8/7/2018 15:46:03, "Baldur Norddahl" 
> wrote:
>
> >Hello
> >
> >There is a lack of bidirectional one fiber (BIDI) options for 40G and
> >100G optics. Usually BIDI is implemented using two CWDM wavelengths,
> >one for tx and one for rx. However there is also a lack of CWDM and
> >DWDM options for 40G and 100G.
> >
> >Would it be possible to use an optical circulator like this one
> >(customized to 1310 nm)?
> >
> >https://www.fs.com/de/en/products/33364.html
> >
> >Combined with a traditional two fiber 1310 nm 10 km 40G QSFP module
> >like this: https://www.fs.com/de/en/products/24422.html
> >
> >The link distance would be 5 km.
> >
> >The optical circulator separates tx and rx by the direction the light
> >travels in. It would work even though both directions use the same
> >wavelength. There will likely be some reflection but hopefully
> >attenuated enough that it is regarded as background noise.
> >
> >Has anyone done this? Any reason it would not work?
> >
> >Regards,
> >
> >Baldur
> >
>
> The main issue you're going to run into (especially trying to plug
> anything into a DWDM shelf) is 40G and 100G transceivers usually emit 4
> lanes of traffic instead of a single lane like 10 and 1G optics do.
>
> I'd imagine that's why there are so few solutions that don't involve
> things like OTN.
>
>

Re: optical circulator as a bidirectional one fiber solution

[Ben Cannon]

Good news about almost all optics, their Rx window is pretty wide. Meaning a 
1550nm optic will activate the receiver on a 1560nm optic just fine (and 
probably anything in the 1500nm band).  Careful use of specialized single 
strand DWDM muxes (FS.com) can yield great bidi-like results with increased 
channel count. 

-Ben

> On Aug 13, 2018, at 10:49 AM, Eric Kuhnke  wrote:
> 
> Something that is broadly the same as a coherent 100G QPSK single wavelength 
> optical module, but in two different frequencies, and a passive CWDM 
> mux/demux prism at each end might work. The limitation would be availability 
> of optics for a modern 100G MSA that are both coherent and Tx/Rx at two 
> different THz frequencies.
> 
> Or with some box and vendor equipment in between, such as: 
> 
> http://cdn.extranet.coriant.com/resources/Application-Notes/AN_Groove_Bidirectional_Fiber_74C0169.pdf?mtime=20180206023321
> 
>> On Tue, Aug 7, 2018 at 1:00 PM Daniel Corbe  wrote:
>> On 8/7/2018 15:46:03, "Baldur Norddahl"  
>> wrote:
>> 
>> >Hello
>> >
>> >There is a lack of bidirectional one fiber (BIDI) options for 40G and 
>> >100G optics. Usually BIDI is implemented using two CWDM wavelengths, 
>> >one for tx and one for rx. However there is also a lack of CWDM and 
>> >DWDM options for 40G and 100G.
>> >
>> >Would it be possible to use an optical circulator like this one 
>> >(customized to 1310 nm)?
>> >
>> >https://www.fs.com/de/en/products/33364.html
>> >
>> >Combined with a traditional two fiber 1310 nm 10 km 40G QSFP module 
>> >like this: https://www.fs.com/de/en/products/24422.html
>> >
>> >The link distance would be 5 km.
>> >
>> >The optical circulator separates tx and rx by the direction the light 
>> >travels in. It would work even though both directions use the same 
>> >wavelength. There will likely be some reflection but hopefully 
>> >attenuated enough that it is regarded as background noise.
>> >
>> >Has anyone done this? Any reason it would not work?
>> >
>> >Regards,
>> >
>> >Baldur
>> >
>> 
>> The main issue you're going to run into (especially trying to plug 
>> anything into a DWDM shelf) is 40G and 100G transceivers usually emit 4 
>> lanes of traffic instead of a single lane like 10 and 1G optics do.
>> 
>> I'd imagine that's why there are so few solutions that don't involve 
>> things like OTN.
>> 

DAZN CDN

[Graham Johnston]

Anyone from DAZN here, or anyone know what CDN is used for their content? I'm 
specifically curious about NFL Sunday Ticket content in case it makes a 
difference.

Thanks,
Graham

Re: optical circulator as a bidirectional one fiber solution

[Eric Kuhnke]

For 1 and 10Gbps OOK modulation yes, but not for something like a ITU DWDM
grid channelized or tunable coherent optic. In which the (QPSK, 8PSK,
16QAM) signal has a specific THz width and frequency not unlike a radio
operating in a very, very narrow waveguide.


On Mon, Aug 13, 2018 at 1:57 PM Ben Cannon  wrote:

> Good news about almost all optics, their Rx window is pretty wide. Meaning
> a 1550nm optic will activate the receiver on a 1560nm optic just fine (and
> probably anything in the 1500nm band).  Careful use of specialized single
> strand DWDM muxes (FS.com) can yield great bidi-like results with
> increased channel count.
>
> -Ben
>
> On Aug 13, 2018, at 10:49 AM, Eric Kuhnke  wrote:
>
> Something that is broadly the same as a coherent 100G QPSK single
> wavelength optical module, but in two different frequencies, and a passive
> CWDM mux/demux prism at each end might work. The limitation would be
> availability of optics for a modern 100G MSA that are both coherent and
> Tx/Rx at two different THz frequencies.
>
> Or with some box and vendor equipment in between, such as:
>
>
> http://cdn.extranet.coriant.com/resources/Application-Notes/AN_Groove_Bidirectional_Fiber_74C0169.pdf?mtime=20180206023321
>
> On Tue, Aug 7, 2018 at 1:00 PM Daniel Corbe 
> wrote:
>
>> On 8/7/2018 15:46:03, "Baldur Norddahl" 
>> wrote:
>>
>> >Hello
>> >
>> >There is a lack of bidirectional one fiber (BIDI) options for 40G and
>> >100G optics. Usually BIDI is implemented using two CWDM wavelengths,
>> >one for tx and one for rx. However there is also a lack of CWDM and
>> >DWDM options for 40G and 100G.
>> >
>> >Would it be possible to use an optical circulator like this one
>> >(customized to 1310 nm)?
>> >
>> >https://www.fs.com/de/en/products/33364.html
>> >
>> >Combined with a traditional two fiber 1310 nm 10 km 40G QSFP module
>> >like this: https://www.fs.com/de/en/products/24422.html
>> >
>> >The link distance would be 5 km.
>> >
>> >The optical circulator separates tx and rx by the direction the light
>> >travels in. It would work even though both directions use the same
>> >wavelength. There will likely be some reflection but hopefully
>> >attenuated enough that it is regarded as background noise.
>> >
>> >Has anyone done this? Any reason it would not work?
>> >
>> >Regards,
>> >
>> >Baldur
>> >
>>
>> The main issue you're going to run into (especially trying to plug
>> anything into a DWDM shelf) is 40G and 100G transceivers usually emit 4
>> lanes of traffic instead of a single lane like 10 and 1G optics do.
>>
>> I'd imagine that's why there are so few solutions that don't involve
>> things like OTN.
>>
>>

RE: optical circulator as a bidirectional one fiber solution

[Jameson, Daniel]

You would still need to frequency shift TX and RX.  They are travelling 
opposite directions on the same piece of glass;  as the traffic rate increases 
the likelihood of collisions increases and you’ll start to get errors.  The 
collision would either cancel the ‘bit’ or act like OBI and get erroneous bits 
and checksum errors or missing chunks from the constellation.  There are BIDI 
100G transceivers for Multi-mode available today based on the Foxconn optics, 
I’d imagine we’ll see BIDI for the O and C bands in the next 18 months.

From: NANOG [mailto:nanog-boun...@nanog.org] On Behalf Of Eric Kuhnke
Sent: Monday, August 13, 2018 3:56 PM
To: b...@6by7.net; nanog@nanog.org list
Subject: Re: optical circulator as a bidirectional one fiber solution

For 1 and 10Gbps OOK modulation yes, but not for something like a ITU DWDM grid 
channelized or tunable coherent optic. In which the (QPSK, 8PSK, 16QAM) signal 
has a specific THz width and frequency not unlike a radio operating in a very, 
very narrow waveguide.


On Mon, Aug 13, 2018 at 1:57 PM Ben Cannon 
mailto:b...@6by7.net>> wrote:
Good news about almost all optics, their Rx window is pretty wide. Meaning a 
1550nm optic will activate the receiver on a 1560nm optic just fine (and 
probably anything in the 1500nm band).  Careful use of specialized single 
strand DWDM muxes (FS.com) can yield great bidi-like results 
with increased channel count.
-Ben

On Aug 13, 2018, at 10:49 AM, Eric Kuhnke 
mailto:eric.kuh...@gmail.com>> wrote:
Something that is broadly the same as a coherent 100G QPSK single wavelength 
optical module, but in two different frequencies, and a passive CWDM mux/demux 
prism at each end might work. The limitation would be availability of optics 
for a modern 100G MSA that are both coherent and Tx/Rx at two different THz 
frequencies.

Or with some box and vendor equipment in between, such as:

http://cdn.extranet.coriant.com/resources/Application-Notes/AN_Groove_Bidirectional_Fiber_74C0169.pdf?mtime=20180206023321

On Tue, Aug 7, 2018 at 1:00 PM Daniel Corbe 
mailto:dco...@hammerfiber.com>> wrote:
On 8/7/2018 15:46:03, "Baldur Norddahl" 
mailto:baldur.nordd...@gmail.com>>
wrote:

>Hello
>
>There is a lack of bidirectional one fiber (BIDI) options for 40G and
>100G optics. Usually BIDI is implemented using two CWDM wavelengths,
>one for tx and one for rx. However there is also a lack of CWDM and
>DWDM options for 40G and 100G.
>
>Would it be possible to use an optical circulator like this one
>(customized to 1310 nm)?
>
>https://www.fs.com/de/en/products/33364.html
>
>Combined with a traditional two fiber 1310 nm 10 km 40G QSFP module
>like this: https://www.fs.com/de/en/products/24422.html
>
>The link distance would be 5 km.
>
>The optical circulator separates tx and rx by the direction the light
>travels in. It would work even though both directions use the same
>wavelength. There will likely be some reflection but hopefully
>attenuated enough that it is regarded as background noise.
>
>Has anyone done this? Any reason it would not work?
>
>Regards,
>
>Baldur
>

The main issue you're going to run into (especially trying to plug
anything into a DWDM shelf) is 40G and 100G transceivers usually emit 4
lanes of traffic instead of a single lane like 10 and 1G optics do.

I'd imagine that's why there are so few solutions that don't involve
things like OTN.

Re: optical circulator as a bidirectional one fiber solution

[Ben Cannon]

What about 100Ghz ITU spacing on the tx, are the rx optics broad enough to take 
the off-band input?

-Ben

> On Aug 13, 2018, at 3:19 PM, Jameson, Daniel  
> wrote:
> 
> You would still need to frequency shift TX and RX.  They are travelling 
> opposite directions on the same piece of glass;  as the traffic rate 
> increases the likelihood of collisions increases and you’ll start to get 
> errors.  The collision would either cancel the ‘bit’ or act like OBI and get 
> erroneous bits and checksum errors or missing chunks from the constellation.  
> There are BIDI 100G transceivers for Multi-mode available today based on the 
> Foxconn optics, I’d imagine we’ll see BIDI for the O and C bands in the next 
> 18 months.
>  
> From: NANOG [mailto:nanog-boun...@nanog.org] On Behalf Of Eric Kuhnke
> Sent: Monday, August 13, 2018 3:56 PM
> To: b...@6by7.net; nanog@nanog.org list
> Subject: Re: optical circulator as a bidirectional one fiber solution
>  
> For 1 and 10Gbps OOK modulation yes, but not for something like a ITU DWDM 
> grid channelized or tunable coherent optic. In which the (QPSK, 8PSK, 16QAM) 
> signal has a specific THz width and frequency not unlike a radio operating in 
> a very, very narrow waveguide.
>  
>  
> On Mon, Aug 13, 2018 at 1:57 PM Ben Cannon  wrote:
> Good news about almost all optics, their Rx window is pretty wide. Meaning a 
> 1550nm optic will activate the receiver on a 1560nm optic just fine (and 
> probably anything in the 1500nm band).  Careful use of specialized single 
> strand DWDM muxes (FS.com) can yield great bidi-like results with increased 
> channel count. 
> 
> -Ben
> 
> On Aug 13, 2018, at 10:49 AM, Eric Kuhnke  wrote:
> 
> Something that is broadly the same as a coherent 100G QPSK single wavelength 
> optical module, but in two different frequencies, and a passive CWDM 
> mux/demux prism at each end might work. The limitation would be availability 
> of optics for a modern 100G MSA that are both coherent and Tx/Rx at two 
> different THz frequencies.
>  
> Or with some box and vendor equipment in between, such as:
>  
> http://cdn.extranet.coriant.com/resources/Application-Notes/AN_Groove_Bidirectional_Fiber_74C0169.pdf?mtime=20180206023321
>  
> On Tue, Aug 7, 2018 at 1:00 PM Daniel Corbe  wrote:
> On 8/7/2018 15:46:03, "Baldur Norddahl"  
> wrote:
> 
> >Hello
> >
> >There is a lack of bidirectional one fiber (BIDI) options for 40G and 
> >100G optics. Usually BIDI is implemented using two CWDM wavelengths, 
> >one for tx and one for rx. However there is also a lack of CWDM and 
> >DWDM options for 40G and 100G.
> >
> >Would it be possible to use an optical circulator like this one 
> >(customized to 1310 nm)?
> >
> >https://www.fs.com/de/en/products/33364.html
> >
> >Combined with a traditional two fiber 1310 nm 10 km 40G QSFP module 
> >like this: https://www.fs.com/de/en/products/24422.html
> >
> >The link distance would be 5 km.
> >
> >The optical circulator separates tx and rx by the direction the light 
> >travels in. It would work even though both directions use the same 
> >wavelength. There will likely be some reflection but hopefully 
> >attenuated enough that it is regarded as background noise.
> >
> >Has anyone done this? Any reason it would not work?
> >
> >Regards,
> >
> >Baldur
> >
> 
> The main issue you're going to run into (especially trying to plug 
> anything into a DWDM shelf) is 40G and 100G transceivers usually emit 4 
> lanes of traffic instead of a single lane like 10 and 1G optics do.
> 
> I'd imagine that's why there are so few solutions that don't involve 
> things like OTN.

Re: optical circulator as a bidirectional one fiber solution

[Brandon Martin]

On 08/13/2018 06:24 PM, Ben Cannon wrote:
What about 100Ghz ITU spacing on the tx, are the rx optics broad enough 
to take the off-band input?


Non-coherent receivers usually seem to even when paired with DWDM grid 
transceivers.  You're paying for the tightly controlled laser on those, 
not so much the receiver.


Coherent of course will demod whatever you tune them to (and pretty well 
reject other stuff) but are therefore subject to their LO tuning range.


Figure your typical OOK optical receiver is roughly like a single-stage 
AM radio receiver (crystal radio or similar, just followed by 
amplification to logic level) without any front-end filtering on it. 
The demux in front of it, where present, is the front-end filter.  It'll 
capture and downconvert to RF basically anything it sees.

--
Brandon Martin

RE: optical circulator as a bidirectional one fiber solution

[Jameson, Daniel]

If we were talking 10G, adjacent channels,  add a TFFL filter it *Should* work. 
 100G isn’t just on-off at a high clock rate,  it’s also modulated  around the 
center frequency,  I don’t think it’d work even with a  wideband receiver.

From: Ben Cannon [mailto:b...@6by7.net]
Sent: Monday, August 13, 2018 4:24 PM
To: Jameson, Daniel
Cc: Eric Kuhnke; nanog@nanog.org list
Subject: Re: optical circulator as a bidirectional one fiber solution

What about 100Ghz ITU spacing on the tx, are the rx optics broad enough to take 
the off-band input?
-Ben

On Aug 13, 2018, at 3:19 PM, Jameson, Daniel 
mailto:daniel.jame...@tdstelecom.com>> wrote:
You would still need to frequency shift TX and RX.  They are travelling 
opposite directions on the same piece of glass;  as the traffic rate increases 
the likelihood of collisions increases and you’ll start to get errors.  The 
collision would either cancel the ‘bit’ or act like OBI and get erroneous bits 
and checksum errors or missing chunks from the constellation.  There are BIDI 
100G transceivers for Multi-mode available today based on the Foxconn optics, 
I’d imagine we’ll see BIDI for the O and C bands in the next 18 months.

From: NANOG [mailto:nanog-boun...@nanog.org] On Behalf Of Eric Kuhnke
Sent: Monday, August 13, 2018 3:56 PM
To: b...@6by7.net; 
nanog@nanog.org list
Subject: Re: optical circulator as a bidirectional one fiber solution

For 1 and 10Gbps OOK modulation yes, but not for something like a ITU DWDM grid 
channelized or tunable coherent optic. In which the (QPSK, 8PSK, 16QAM) signal 
has a specific THz width and frequency not unlike a radio operating in a very, 
very narrow waveguide.


On Mon, Aug 13, 2018 at 1:57 PM Ben Cannon 
mailto:b...@6by7.net>> wrote:
Good news about almost all optics, their Rx window is pretty wide. Meaning a 
1550nm optic will activate the receiver on a 1560nm optic just fine (and 
probably anything in the 1500nm band).  Careful use of specialized single 
strand DWDM muxes (FS.com) can yield great bidi-like results 
with increased channel count.
-Ben

On Aug 13, 2018, at 10:49 AM, Eric Kuhnke 
mailto:eric.kuh...@gmail.com>> wrote:
Something that is broadly the same as a coherent 100G QPSK single wavelength 
optical module, but in two different frequencies, and a passive CWDM mux/demux 
prism at each end might work. The limitation would be availability of optics 
for a modern 100G MSA that are both coherent and Tx/Rx at two different THz 
frequencies.

Or with some box and vendor equipment in between, such as:

http://cdn.extranet.coriant.com/resources/Application-Notes/AN_Groove_Bidirectional_Fiber_74C0169.pdf?mtime=20180206023321

On Tue, Aug 7, 2018 at 1:00 PM Daniel Corbe 
mailto:dco...@hammerfiber.com>> wrote:
On 8/7/2018 15:46:03, "Baldur Norddahl" 
mailto:baldur.nordd...@gmail.com>>
wrote:

>Hello
>
>There is a lack of bidirectional one fiber (BIDI) options for 40G and
>100G optics. Usually BIDI is implemented using two CWDM wavelengths,
>one for tx and one for rx. However there is also a lack of CWDM and
>DWDM options for 40G and 100G.
>
>Would it be possible to use an optical circulator like this one
>(customized to 1310 nm)?
>
>https://www.fs.com/de/en/products/33364.html
>
>Combined with a traditional two fiber 1310 nm 10 km 40G QSFP module
>like this: https://www.fs.com/de/en/products/24422.html
>
>The link distance would be 5 km.
>
>The optical circulator separates tx and rx by the direction the light
>travels in. It would work even though both directions use the same
>wavelength. There will likely be some reflection but hopefully
>attenuated enough that it is regarded as background noise.
>
>Has anyone done this? Any reason it would not work?
>
>Regards,
>
>Baldur
>

The main issue you're going to run into (especially trying to plug
anything into a DWDM shelf) is 40G and 100G transceivers usually emit 4
lanes of traffic instead of a single lane like 10 and 1G optics do.

I'd imagine that's why there are so few solutions that don't involve
things like OTN.

Re: optical circulator as a bidirectional one fiber solution

[Jared Mauch]

> On Aug 13, 2018, at 5:56 PM, Eric Kuhnke  wrote:
> 
> For 1 and 10Gbps OOK modulation yes, but not for something like a ITU DWDM 
> grid channelized or tunable coherent optic. In which the (QPSK, 8PSK, 16QAM) 
> signal has a specific THz width and frequency not unlike a radio operating in 
> a very, very narrow waveguide.
> 

There are some DCO 100G coherent optics on the market, but I think this thread 
is more about why there’s not much in the way of 40/100g transmission speed, 
but it really is about how 10G was one of the last walls where OOK was a thing. 
 Once you went 40G/100G you went to parallel signals, either in CDWDM form or 
parallel signals on parallel fibers (eg: those MPT/MPO cables we all get to 
enjoy).

If you talk to the optics folks you can get customized tx/rx optics, and in 
theory you could get an optimized 100G-LR4 or 100G-ER4-Lite optic, but even 
those the reach is quite limited to around 25Km while in the 10G space you can 
get up to 80-120km with a filter and a bunch of pluggable fixed channel (or 
tunable) optics.

You can get decent 100G OEO solutions for a fair price per channel but you’re 
still looking at $$ compared to just building a 40x10G system due to the cost 
differences.

Without more details like your link budget, etc.. it’s hard to suggest options 
other than get at least 2 fibers :-)

I know some vendors are working on some creative 1 fiber solutions due to the 
way the dark fiber is taxed in Europe, so you should be talking to the usual 
suspects..

- Jared