Best components for a full mvno core network?

Mon Oct 21 13:13:00 UTC 2019

Hello Javier,

Well, if we take a step back to goals, I would like first to point that
going Full MVNO might not be the best solution for us (roaming alone seems
like quite a hassle, not to mention handsets management).

My focus here is narrower, as I am mostly trying here to assert what the
possibilities for the core are, and if there are reasonable alternatives to
the fully integrated solutions of the big providers.

That being said, I am not sure how our specific goals here would impact
much the architecture: aren’t there a lot of constraints due to the 3GPP
requirements?  It seems to leave little room for creativity.

To provide a bit of context and answer you:

We are historically providing solutions on fixed networks, with a strong
bend toward business end-users. We are also used to have a lot of control
over our architecture, most of our services running over open-source and/or
in-house solutions.

Being able to provide our services on mobile accesses is now a necessity.
For this we already are light MVNO, using two different MNOs. Thanks to
forced routing, it mostly does the job regarding voice. Data could be
managed also. SMS is proving trickier.

But each MNO have their own products set: building offers that would work
on both is tedious and necessitate compromises that tend to make our
marketing people unhappy. Not to talk about supporting two provisioning
chains, two SIMs supply chains, etc… These problems would only get worse if
we add other MNOs to the mix.

We are also stuck with the roadmap of the MNOs (VoLTE and VoWifi are but
distant “maybe later” possibilities).

So, in one word, this is about autonomy. And its cost.

Regards,

Dario Renaud

Le ven. 18 oct. 2019 à 17:44, Javier J <javier at advancedmachines.us> a
écrit :

> This is interesting but so many variables to unpack to determin what the
> right solution is. What are the main goals of your org? What exact pain
> points are you trying to fix?
>
>
>
> On Wed, Oct 16, 2019, 8:28 AM Dario Renaud <dario.renaud at gmail.com> wrote:
>
>> Hello,
>>
>> At my day job, we are considering going Full MVNO. Which means building a
>> mobile core network.
>>
>> I was wondering if some of you would have feedback or advices on the
>> solutions currently available?
>>
>> We would like to avoid the big providers (Ericsson & such).
>> Ideally, something opensource, or, if proprietary, a company maybe
>> willing to license access to the code (one can dream).
>>
>> There seems to be a lot of bits and pieces available out there, with a
>> mix of full, fullish or partial solutions. This makes for quite the puzzle.
>>
>> Among the ones I found most interesting:
>>
>> nextEPC, covering, well, the EPC… (https://github.com/nextepc/nextepc).
>> It looks like the more active open EPC implementation out there.
>>
>> And it seems that Yate people have a commercial product covering
>> basically everything needed (
>> https://yatebts.com/solutions_and_technology/mobile_virtual_network_operator/).
>>
>>
>> What do you think?
>>
>> Regards
>>
>> Dario Renaud
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nanog.org/pipermail/nanog/attachments/20191021/3e881e82/attachment.html>