Why PRIs Will Die In 2014
It's 2014, and UCC is going mainstream. SIP trunking is gaining major interest and is being offered by most, if not all, major carriers at this juncture. Primary Rate Interface (PRI), a form of T-1 digital service running at 1.544Mb, 23 channels (and 23 D and 1 B channel), has been implemented now for over 10 years and has become the defacto standard for any site with 50+ users and above. Multiple PRIs have been implemented with hunting among several to create what can appear to be a much "larger" circuit. These "pigtailed" PRIs are common larger sites and in Contact Centers especially.
For most of us who have been around for at least the last 10+ years, we remember the days when it took upwards of up to 4-8 hours to 'turn up' a PRI as stable and fully functioning and then release it into a full production environment. Fast forward to SIP trunking - for those of us who have implemented such, a nearly identical path has followed. That's not to say SIP trunking does not work - once implemented, it can be a game-changing technology. Within our consulting practice we have realized that SIP trunking is complex and have developed detailed test plans for SIP trunking delivery, SBC integration and programming, developed designs for redundancy, failover, implemented CODECs (G.711, G.729, and even G.722 HD CODEC), number portability, and more. Yes, SIP trunking is complex, but in the end the benefits can far outweigh the difficulties implementing such. Our evaluations and procurements for SIP trunking have delivered savings between 20% - 60%+ annually as compared with legacy PRI installs.
The Death of PRIs As We Know It
From my perspective, I believe we will see the beginning of the death of PRIs, as we know it, in 2014. This is not to say they will go away in 2014, rather the phasing out of this legacy technology over the next couple of years. Why? For several key reasons:
36 Month Carrier Cycle. In the carrier world, historically older technologies are replaced by either better price per Mb/price per mile and/or newer technologies that are more robust, provide more bandwidth, and lower rates compared with the older technologies. It takes about 36 months for such a cycle, and it is subtle and not obvious to many of us. For example, X.25 packet switching evolved to Frame Relay which evolved to a QoS-based Frame Relay, aka MPLS, which is now being replaced with Layer 2 Ethernet in many cases on data networks. SIP trunking is a form of MPLS under the covers.
It also happened with T-1s years ago, with legacy tie lines and off-premise extensions, now replaced with 2-point or multi-point MPLS circuits/networks at a cost far less than a traditional point-to-point tie line that carrier geographic mileage costs. From a contract perspective, we typically see a price drop on currently implemented networks at month 36 (time of renewal), sometimes significant. From a cycle point of view, it's time for the carriers to abandon the older PRI technology over SIP trunking, which is exactly what is taking place today.
PRIs Are Digital. PRIs are a legacy technology, period. With everything now going IP, why does one require a gateway to convert a signal from digital to IP on a VoIP/UCC infrastructure? In the world of SIP trunking gateways are no longer required.
PRIs Are More Expensive Than SIP Trunking. In almost every case, we have seen SIP trunking costs outperform PRI trunking. Some carriers are even offering semi-flat rate or bundled plans utilizing SIP trunking. In addition, SIP trunking can be layered on top of an existing Internet circuit with QoS, thus savings can be even greater. For a 500 station healthcare client of ours, we were able to increase bandwidth by a factor of 10x for MPLS and Internet, and use a portion for a PRI equivalent via SIP trunking, and was able to create a failover environment to a second site, all for less cost than the current legacy model, which was seriously affected after Hurricane Sandy.
Carrier Sales Teams Get Comp'd Better For Selling New State-Of-The-Art Product, Not Legacy Products. Carriers incentivize their sales teams and VARs based on newer technologies, and I predict that incentives for PRIs will diminish this coming year, being replaced by SIP trunking. Expect in 2014 a diminishing of internal and channel partner compensation packages selling PRI services.
Carriers No Longer Want To Be In The Legacy Business. For similar reasons to the above statement, carriers continuously need to look at margin and consider what are the best technologies that require less maintenance, less hardware, and easier to monitor. SIP trunking provides all of these advantages.
SIP Trunking Is More Redundant Than PRIs. SIP trunking, when designed right, can provide up to 2-3x the bandwidth and conversation paths than PRI technology over multiple geographies while still reducing annual costs, thus creating a more redundant topology and additional failover capabilities. This provides a more redundant environment for Disaster Recovery purposes.
SIP Trunking Is A Perfect Venue For Federating. As SIP trunking is IP and QoS ready, SIP trunking can be used for off-site remote users, softphones, mobile users, and Federated partners.
SIP Trunking Is Video-Ready. SIP trunking, together with SBCs, can be leveraged for QoS designed videoconferencing, different than PRIs with ISDN dial up capability. H.323, H.264, and T.120 CODECs are available in SBCs for video-over-SIP, including running video over SIP at 1Mb bandwidth HD purposes not available using PRI technology.
Session Border Controllers are about providing firewall capability. At a high level these toolsets include NAT/PAT, Encryption - IPSec, TLS, SSL, etc., DOS and DDOS, ACLs, Application Security, Logging, Monitoring, and Diagnostics.
SBCs also prevent SIP specific overload conditions, provide RTP ports in sync with SIP signaling, track session state and provide uninterrupted service, scale to hundreds/thousands of real-time sessions, and solve multi-vendor SIP interoperability problems. Yes, all the carriers provide SBCs, but do not guarantee the protection required by an enterprise. Gartner Group recommends enterprise-side SBCs as I do in my consulting practice. So in my opinion, it's simply not good enough to rely on a carrier to prevent attacks and interoperability issues without your own SBCs.
Audiocodes, for one, is seeing more and more interest in the SIP trunking and thus the SBC market. They have always been a 'player' in small-to-medium capacity SIP trunking market, and recently announced a new high capacity SBC. In my opinion, it's in response to the greater demand for larger SBCs requested by enterprises. Noting that an increasing number of large enterprises are using centralized SIP trunking services to consolidate their calling traffic, AudioCodes is looking to provide solutions in this larger space.
Audiocodes has introduced their Mediant 9000 SBC, which can support up to 16,000 simultaneous sessions. This now allows Audiocodes to compete head on with more established larger SBC providers, including Acme Packet/Oracle, Sonus, Cisco and GENBAND among those in the larger SBC space. The Mediant 9000 also includes Microsoft Lync toolsets and configurators, and session management to integrate multiple PBX manufacturers for facilitating dialing plans and 'integration' among multiple manufacturers.
It's always better to be proactive vs. reactive, in my opinion, SIP trunking will begin to replace PRIs over time, spelling the death of PRIs as we know it today. Sure PRIs will still have their niche for back up to SIP trunking and for places where SIP is not available now or in the near future. But to plan on PRIs and not consider SIP trunking for a green field opportunity or possible replacement as carrier contracts come due for renewal is unwise at best. So it's time to plan to retire PRIs and include SIP trunking and SBCs (AudioCodes Mediant series for one) as a part of your 2-3 year enterprise planning strategy. And include voice, video, federating, and remote and mobile workers as a part of your strategy. Start with voice today, but plan for more tomorrow.