Written by Chief Technology Officer, Josh Reinmann
As PBX’s have grown throughout the years, the need for a more flexible, feature rich and scalable connection to the PSTN has grown with it. T1’s and E1’s became the standard of connectivity in the early 1990’s, growing from un-channelized DS1’s to full featured PRI’s including Direct Inward Dial and Caller ID services. However, a T1 is limited in capacity, redundancy capabilities and changes often require new installations or downtime.
As the PBX’s themselves advanced and VoIP became the standard in the early 2000’s, PSTN connectivity needed an upgrade as well. In comes Session Initiated Protocol (SIP), as an IP-based protocol for voice and video. The advantages of SIP were apparent from the start, though many were uncomfortable with the change and slow to adapt. Over time, these advantages could no longer be ignored. The most prominent benefits of migrating to SIP include scalability, expanded redundancy capabilities (as compared to T1’s), improved service delivery and turn up, and security.
SIP trunks are highly scalable and can be configured as such in multiple ways. First, where a PRI was limited to the T1 on which it was delivered (or T3 for larger environments), maxing out at 23 voice channels before requiring an additional physical circuit, SIP trunks can be delivered in multiple forms including data T1’s, MPLS circuits, private ethernet circuits (MPLS, P2P, etc) and even directly over the internet. The scale is limited by the circuit capacity itself. A 10mb circuit (small by today’s standards) can handle up to 320 calls using the g.729 codec for compression where a similarly-sized T1 group can only handle 138 concurrent calls. In addition, the circuit doesn’t need to be dedicated to voice. Since SIP is an application running on IP, it can be comingled with data; this is revolutionary. The Session Border Controllers (SBC), the interface between the carrier and PBX, grew alongside SIP trunks. Enterprise-grade SBC’s can handle 12,000 or more concurrent calls on a single device; SBC’s can even be virtualized, something which was inconceivable for PRI’s, further increasing the flexibility. This great ability to centralize the service and even share the physical medium with other services (e.g. internet and WAN) can greatly reduce costs both in hard dollars for the monthly service and in personnel regarding the operational time required to manage multiple PRI’s and their associated equipment.
While PRI’s had redundant capabilities, they were often limited to the single site to which they were installed or, if you chose the right carrier, the city. Enter SIP. With SIP, not only can you gain redundancy across hardware using redundancy protocols (e.g. HSRP) with no loss of calls during an outage (not achievable with PRI’s), you can a;sp get redundancy across geographies limited only by the laws of the nations across which you require service. This leads to an amazing ability of centralized SIP. Along with a robust WAN, an enterprise with multiple locations can now have their PSTN connectivity centralized into two datacenters. This offers redundancy, simplicity and savings as there is no need to support hardware and circuits in every site.
Service Delivery and Turn Up
Prior to SIP, ordering new circuits was a costly and time-consuming process including the installation of new hardware and reconfiguration of trunk groups which carries a small amount of risk. With SIP trunks, expanding capacity is as easy as a service request to increase the concurrent call paths (CCP) assuming there is sufficient bandwidth on the circuit which there almost always is. Due to the ease of increasing service capacity, the time to change the service is drastically reduced from a typical 90 days for PRI’s to normally 2 weeks with SIP. In addition, to the ability to turn up new service can be very quick and easy because SIP can also be delivered over the internet in a pinch.
PRI’s were considered fairly secure in their day due to the copper on which they ran going directly back to the central office (CO) leaving little room for compromise. However, requiring more PRI’s means the exposure to risk is increased. With SIP being an application running on an IP network, there are more areas for compromise, however, using centralized SIP greatly reduces the attack plane. Additionally, with proper configuration and both modern and future encryption capabilities, we’re able to secure SIP for today’s requirements and beyond. Lastly, SIP trunks can be protected by firewalls and monitored for malicious activity using network monitoring tools.
In the end, SIP is now king. The ability to deliver centralized SIP, at scale, is a very big reason for its success. The reduced complexity matched with a reduced cost makes it more attractive with organizations around the world. Security is not much of a concern. The flexibility to add new features and scale with the “flip of a switch” means SIP be here for a long time to come. Furthermore, as the types of connections to our datacenters and cloud solutions continue to evolve, the ability to decouple the physical medium from the IP service on which SIP is based, gives it staying power. For the final nail in the coffin, almost all PRI service delivered in the US today is delivered as SIP and then converted to PRI locally making PRIs all but obsolete.
If you haven’t already converted to centralized SIP, what are you waiting for?
If you have any questions about centralized SIP, please contact Fidelus Technologies at 1.866.343.3587 or email firstname.lastname@example.org.
Get in touch with Fidelus today to learn more.