[SR-Users] Replacing an ACME Packet Net-Net SBC

Stefan Sayer stefan.sayer at googlemail.com
Fri Feb 28 16:15:55 CET 2014


I'm not going to comment on the relative merits of commercial products
like Net-Net SBC and ABC SBC, as this is an open source focused list
(we may continue this on kamailio-business if interested; of course
I'd also have a few things to say e.g. regarding HA, DOS or interop,
and we are more than happy to take into account any suggestion on how
to improve the product).

But relevant to the open source community is that we (FRAFOS) are
building the ABC SBC on top of the open source SEMS SBC software, thus
a remarkable part of the tech lands finally in the open source
version, for example off the top of my head in the last months:
- registration caching, reg throttling
- multipe routing target options, DNS SRV improvements, failover,
blacklisting etc
- many multi interfaces/bridging/if force options
- RTP rate limits
- hold handling options
- tcp transport
- extended call control interface supporting switch/pbx style call
flows, with DSM scripting
- lots of performance improvements and tuning options

Granted, all this is only in git master, the 1.6 release has
unfortunately no yet materialized and some know-how, time and custom
development is necessary to use all of that. But still it's free and
open source: take it, use it, look at the source, modify it, build on it.

Enough ad of a different open source project (I hope this is OK
though, remembering that SEMS and SER have the same origins)...

Best Regards

o Luis Silva on 02/28/2014 10:21 AM:
> Hi Melanie,
> Now I'm curious about which tests are you mentioning. Also, Acme
> Packet/Oracle as many different SBCs models, which can be selected
> according with the customer's requirements. 
> When talking about the infinite number of features I wasn't really
> thinking in H.323. Although, probably there were customers like you a
> couple of years ago that thought they wouldn't need some features and
> regret they didn't pick Acme Packet, after they received new
> requirements on their networks. Acme SBC can also route using any SIP
> message element. Just use an HMR and according with what you want,
> just add a Route header (as simple as that... :) ). 
> Regarding traffic shaping, you can also do it based on media-profiles.
> Comparing traffic shaping with Acme Packet DoS protection doesn't make
> much sense. Their DoS is one of the features that made their SBC what
> it is now, the leader worldwide. And do you know what it means? It is
> tested against any tool out there performing attacks (more customers
> means more testing). Can you say the same about the ABC demo?
> One additional thing is that you can for sure say that when deploying
> a Acme Packet SBC you won't have issues interoperate with other
> vendors. This is not marketing material, again this is based on the
> number of customers/deployments that exists worldwide.
> Not to mention real HA (out-of-the-box), simple configuration (yeah,
> I've already implemented an P-CSCF using Kamailio and it was a
> mess...), transcoding, LI, etc.
> Hope it helps,
> Luis
> On Thu, Feb 27, 2014 at 5:33 PM, Melanie Pietersen
> <melanie.pietersen.ml at gmail.com
> <mailto:melanie.pietersen.ml at gmail.com>> wrote:
>     On 2/27/14 10:29 AM, Luis Silva wrote:
>         Don't understand how is it possible to compare Kamailio with
>         the Acme Packet SBC. Just to give an example, the DoS
>         mechanism available on the AP SBC can't be
>          compared with any other solution available on the market. You
>         will also have HMRs, SIP Routing options out-of-the-box (like
>         time of the day routing, sip
>         method based, cost based, traffic classification based, lb,
>         trunk group, enum,lrt, multistage, route header based,
>         redirect, etc etc) and infinite number of
>         features that you will for sure miss if you plan on making
>         that replacement.
>     Hi Luis,
>     the acme product seems actually underperforming compared to
>     Wladimir's SBC demo and measured by the technical capabilities I
>     am interested in.
>     What continues to confuse me is the apparent mixture of technical
>     and marketing terms. The infinite number is respectably long but
>     it includes  many features I will not need like H.323 or appear to
>     be marketing. For example it appears that ABC demo can route by
>     any SIP message element, which is clearly superior and necessary,
>     especially in a deployment with proprietary header fields. Even if
>     it does not make such a long element-by-element feature list :).
>     Similarly it includes several types of traffic shaping, which
>     seems little different from the DoS protection offered by former
>     Acme. Acme's multistage lookup you mentioned seems to be just a
>     fix to quite imperfect design which didn't anticipate cascaded
>     routing logic. I mean I am a little bit careful about assessing
>     technological supremacy based on marketing material.
>     Is there possibly some truly technical-based comparison? I tried
>     to look it up in the archive but did not find some.
>     Melanie
>     _________________________________________________
>     SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing
>     list
>     sr-users at lists.sip-router.org <mailto:sr-users at lists.sip-router.org>
>     http://lists.sip-router.org/__cgi-bin/mailman/listinfo/sr-__users
>     <http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users>
> _______________________________________________
> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
> sr-users at lists.sip-router.org
> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users

More information about the sr-users mailing list