Hello Everyone,
I can only imagine how many times this question has come up since post 2008. Please forgive my reoccurring of the issue.
We are looking to provide carrier grade sip services to our clients world wide. What we need is a lightweight, robust and scalable solution that will allow us to terminate sip calls to our different carriers. Performance, and high throughput are factors very important to my employer. Features such as caller authentication, database back-end, load balancing, and interoperability with asterisk are things we are interested in, as was offered using OpenSER.
With three+ open source proxy servers available on the net puts us in a situation where we have more solutions to choose from, at the same time wish the features from one were available in the other, and vice versa.
With this in mind, we will have to fall back to other factors such as the most reliable, proven and active projects. As mentioned, we would choose functional stability over endless features that we will never use and that add to the projects fingerprint...
I understand that all three projects are forks from OpenSER, people would naturally like to know what differentiates one from the other.
Thanks in Advance,
Nick Khamis Toronto Hydro Telecom
Hi,
have you checked the "sip:providerCE/PRO" solution from sipwise (www.sipwise.com)? Pretty nice solution based on Kamailio and SEMS.
It is worth trying it! -Vlada B.
On 08/14/2011 11:33 PM, Nick Khamis wrote:
Hello Everyone,
I can only imagine how many times this question has come up since post 2008. Please forgive my reoccurring of the issue.
We are looking to provide carrier grade sip services to our clients world wide. What we need is a lightweight, robust and scalable solution that will allow us to terminate sip calls to our different carriers. Performance, and high throughput are factors very important to my employer. Features such as caller authentication, database back-end, load balancing, and interoperability with asterisk are things we are interested in, as was offered using OpenSER.
With three+ open source proxy servers available on the net puts us in a situation where we have more solutions to choose from, at the same time wish the features from one were available in the other, and vice versa.
With this in mind, we will have to fall back to other factors such as the most reliable, proven and active projects. As mentioned, we would choose functional stability over endless features that we will never use and that add to the projects fingerprint...
I understand that all three projects are forks from OpenSER, people would naturally like to know what differentiates one from the other.
Thanks in Advance,
Nick Khamis Toronto Hydro Telecom
SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list sr-users@lists.sip-router.org http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
On Monday 15 August 2011, Nick Khamis wrote:
I can only imagine how many times this question has come up since post 2008. Please forgive my reoccurring of the issue.
Hi Nick,
actually there has been not that much discussion about it recently.
We are looking to provide carrier grade sip services to our clients world wide. [..] With this in mind, we will have to fall back to other factors such as the most reliable, proven and active projects. As mentioned, we would choose functional stability over endless features that we will never use and that add to the projects fingerprint...
In my experience ohloh is really helpful in getting an broad overview about a project, if its healthy or not. E.g.: http://www.ohloh.net/p/sip-router
You can see statistics about the developer community, repository activity. Then have a look to the developer list of the projects, is most of the work done from employees of one company or is this a more distributed effort, and things like this. Have a look to upcoming releases also gives you informations about the development, e.g.: http://sip-router.org/wiki/features/new-in-devel
With regards to the stability the project management is one important aspect, how many and what kind of companies are present there, how big are the deployements etc..: http://www.kamailio.org/w/management/
Maybe have a look to past presentations of the projects as well: http://www.kamailio.org/events/
I understand that all three projects are forks from OpenSER, people would naturally like to know what differentiates one from the other.
Maybe there is a small misunderstanding, they are more or less only two projects in this regards now. The sip-router provides a common repository from that the kamailio (and also ser) can be build. IMHO most people use the kamailio "flavour", though. If you're interested in the details, here are some history informations: http://sip-router.org/history/
Best regards,
Henning
Thank you all for your responses,
Vladimir: Never looked into sipwise, thank you for the link. The only problem is there Linux HA capabilities is not offered in CE, only PRO...
Alex: Nice documentation based on Kamailio! If we choose this product it will be a perfect reference point.
Henning: Just coming out of university as a research student, causes me to have a 'DUH' moment. Of course! Oholoh.... I use to tell other students about it when trying to pull up information regarding test data for ArgoUML and Eclipse (Java projects commonly used in experiments....). Totally slip my mind. In terms of Sip Router vs. Kamailio, is it safe to say the former is a subset of the other? I.e., both function on the same core that provide SIP trunking, proxy and termination services, and Kamailio contains additional features such as B2B etc...?
In terms of clustering, high availability, and more importantly load balancing, what is offered across the board. In terms of "made to scale" this is extremely important to us.
Thanks in Advance!
Cheers,
Nick
On Monday 15 August 2011, Nick Khamis wrote:
[..] Alex: Nice documentation based on Kamailio! If we choose this product it will be a perfect reference point.
Hi Nick,
was the response from Alex also on the list? I somehow missed it.
[..] In terms of Sip Router vs. Kamailio, is it safe to say the former is a subset of the other? I.e., both function on the same core that provide SIP trunking, proxy and termination services, and Kamailio contains additional features such as B2B etc...?
Both are actually the same codebase, the only difference is in packaging, some defines and daemon/script names etc.. http://sip-router.org/releases/ But its the other way round, kamailio is a subset of sip-router.
In terms of clustering, high availability, and more importantly load balancing, what is offered across the board. In terms of "made to scale" this is extremely important to us.
Especially if you look to TCP and TLS there were extensive optimizations and refactorings recently done, e.g. have a look to this benchmark: http://www.kamailio.org/w/2011/05/green-voip-energy-efficiency-and-performac...
With regards to the "build to scale", for example in our backend we use Kamailio to operate services for more than 3 million customers and more than 1 billion minutes/month, so scaling and performance are of course really important for us. Daniel will present in two weeks at the 10 year SER event in Berlin some more interesting usage statistics, I think.
Best regards,
Henning
Hello,
first just some important clarifications regarding the origin. The __first ever__ project in this series was SIP Express Router (aka SER), developed from 2001 inside FhG Fokus Institute, in Berlin, Germany -- as a matter of fact we celebrate 10 years very soon: http://sip-router.org/10-years-ser/.
All the other projects with different names are forks (or were forks since Kamailio and SER are again the same). Hopefully, a diagram that I made will help you get properly the evolution over the time, see second slide in this presentation: http://asipto.com/u/42
OpenSER was started as a fork of SER in 2005. Kamailio was chosen as the new name of OpenSER in July 2008 due to trademark issue, days after some guys decided to fork and created the other project -- how and why is history by now and I don't want to get into it (web site on sourceforge.net (see date of registration) or wikipedia for openser show the markers of the true timelines).
Regarding the present, Kamailio and SER are the two SIP server applications released from the same source code -- there are two since during the 3 years of parallel development (2005-2008) each one built different database structures to keep user profiles and routing data. Because of that you will see some modules with same name in folders modules_k/ and modules_s/. Packaging Kamailio or SER will select only one of the duplicated modules, but when building from sources all of them are present. The development portal (hosting the source code GIT repository, bug tracker, mailing lists, a.s.o.) is sip-router.org
The development is done by the same group of people (since 2008, Kamailo and SER devels are one team), quite large and expanding right now (three new in the past months), all working on the same code base, each contributing to own areas of interest.
By the fact of sharing many developers, I can add that our eco-system includes also the OpenIMSCore project (an open source IMS prototyping system - http://www.openimscore.org - currently integrated in main GIT repository) and SIP Express Media Server (aka SEMS, an open source SIP B2BUA/Media Server - http://www.iptel.org/sems).
Other than that, you can figure out yourself our evolution in the past years reading the release notes of versions 1.5.x, 3.0.x and 3.1.x, the next one 3.2.0 to be done soon: * http://sip-router.org/wiki/features/new-in-devel * http://www.kamailio.org/w/kamailio-openser-v3.1.0-release-notes/ * http://www.kamailio.org/w/kamailio-openser-v3.0.0-release-notes/ * http://www.kamailio.org/w/kamailio-openser-v1.5.0-release-notes/
Last I want to add is that with Kamailio we have a clear release policy, packaging new stable branches every 8-10 months in average, while SER is not that much into packaging lately.
Cheers, Daniel
On 8/15/11 4:33 AM, Nick Khamis wrote:
Hello Everyone,
I can only imagine how many times this question has come up since post 2008. Please forgive my reoccurring of the issue.
We are looking to provide carrier grade sip services to our clients world wide. What we need is a lightweight, robust and scalable solution that will allow us to terminate sip calls to our different carriers. Performance, and high throughput are factors very important to my employer. Features such as caller authentication, database back-end, load balancing, and interoperability with asterisk are things we are interested in, as was offered using OpenSER.
With three+ open source proxy servers available on the net puts us in a situation where we have more solutions to choose from, at the same time wish the features from one were available in the other, and vice versa.
With this in mind, we will have to fall back to other factors such as the most reliable, proven and active projects. As mentioned, we would choose functional stability over endless features that we will never use and that add to the projects fingerprint...
I understand that all three projects are forks from OpenSER, people would naturally like to know what differentiates one from the other.
Thanks in Advance,
Nick Khamis Toronto Hydro Telecom
SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list sr-users@lists.sip-router.org http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
-
Daniel-Constantin Mierla -
Henning Westerholt -
Nick Khamis -
Vladimir Broz