New SIP stack

Simon Woodhead

Simon Woodhead

2nd January 2013

Our new SIP infrastructure is available for your traffic in our Slough site. Manchester will be available shortly and London will follow. The features are below but if you’d like to try it, please request interop information. Of course, if you are not yet a customer you will need to first open a test account.

As you may know, we own our own business logic and enjoy immense agility through the use of Open Source software solutions. However, there are certain things that are just better done in hardware, taking advantage of silicon specifically designed for purpose. Our network runs entirely on Brocade for example with all routing in hardware and similarly our SS7 interconnects run on Dialogic with hardware based DSPs. Hardware gives predictable performance and rock-solid stability in a physically dense way; software gives flexibility and dynamism. This update is a top-to-bottom rebuild which takes the best elements of open-source software agility and fuses them together with hardware based stability and performance, with our own application layer providing the glue in between.

We’re very excited to share this with you and really look forward to you trying it.

Key features are outlined below. For further information please click through.


All business logic and all relevant information is held 100% in RAM for lightening quick operations. This has improved our already good PDD further and we’ve been able to use some of the time saved to add more features.


We have added several limit settings for concurrent calls and the rate of calls for your protection. Of particular interest is calls to known fraud-hotspots where defining a limit on the calls per 12 hours could dramatically limit the pain should you or a customer of yours be compromised.+ MORE

TLS and SRTP now supported.  MORE

Customer specific white- and black- lists. Rather than just being country codes, these work more like an LCR with longest code matching for granular permissions.+ MORE


By popular demand you can now switch service levels / rate decks using a prefix. MORE

You may now configure trunks under your account, each with their own username/password. All account level settings mentioned here are configurable per trunk.+  MORE


UK Directory Enquiries, 999 and access codes (e.g. Speaking Clock) are now individually enabled.+

You may pass a header with outbound calls which will be given back in CDRs making per call reconciliation possible.*  MORE

Trunk IDs will be passed through in CDRs.*


A decision is made on media handling per call-leg. We always proxy media in order to give you stable performance but your codec choice and how it is leaving our network (SS7 or VoIP) will affect where we do it. If coming/going via SS7 media will flow directly to/from our SS7 gateways and be handled in hardware DSPs unless you request HD (or the new Opus codec) when it will be handled on entry to the network, in hardware where possible.


We have further improved the ability for each site to function in the event of others being unavailable and to gracefully inter-work in the event of a local failure.

There is no dependence on a traditional database and we have flattened our n-tier application stack into self-contained business-logic appliances.

We’ve fixed a few bugs and known issues with our legacy proxies – mostly relating to the over-use of generic error codes.

We really hope you like the new features and look forward to your feedback. The new stack is defined as a ‘beta’ but should be stable to include in production routing with our production set-up as a backup. Areas considered ‘beta’ are the new features which may not work as expected.

If you would like to try the new stack, please request interop information. Naturally, if you do not yet have an account you will need to open a test account before being able to participate in the beta.

+ this feature is pending API support but can be enabled for the beta by making a support request.* this feature is pending API support

Related posts