True Voice Resiliency with Cyclix: Better VoIP Reliability & Global Portability through SIP Trunking

Figure 2 is a high-reliability MPLS network where all VoIP telephony is brought into a single ingress/egress point at the headquarters of an enterprise. This is terminated into an IP-PBX that serves and controls smaller branch locations, which just have VoIP phone extensions. The attribute of VoIP that allows this type of faulty network architecture can be used to solve the problem. VoIP service providers like Cyclix have the ability to route multiple ingress/egress traffic into any VoIP signaling point, thus avoiding a single point of failure issue. IP-PBX manufacturers are creating resilient PBXs whereby two IP-PBXs can be deployed to share telephone operations across the distributed enterprise.

Let’s looks at an example to see how bad this architecture could be for a distributed organization in Figure 2.

Company XYZ has its headquarters in Brownsville, Texas. It has 47 smaller remote stores spread out across each state in the continental USA. It is tied to each of the remote branches through a high-reliability MPLS network, managed by one of the top national tier-1 carriers.

All VoIP telephony is brought into a single ingress/egress point at the Texas headquarters, and terminated into an IP-PBX that serves and controls the smaller branch locations. Furthermore, many of the smaller office locations do not have IP-PBXs, but rather just VoIP phone extensions from the main IP-PBX at headquarters.

Along comes a natural disaster. Let’s see what happens. Brownsville is hit by a major hurricane. In the process, Company XYZ ends up with its telecom equipment room 6 feet underwater. What just happened…?

Well, certainly Company XYZ has just lost its communications at headquarters. However, it has now also lost all communications for all of its branches across the nation. So Company XYZ has now lost all its telephony services for all its locations across the entire country, even though there are no natural disasters anywhere else in the country. So ponder what just happened here…

This type of catastrophic VoIP network is being deployed throughout the world “now” to save on communications costs. In the end, no costs are saved with this solution.

By stepping back and analyzing this scenario, we see that this need not happen. The very attribute of VoIP that allowed this faulty type of network to be built in the first place can also be used to solve the problem. Specifically, because VoIP, and VoIP Service providers like Cyclix have the ability to route ingress/egress traffic to any VOIP signaling point, multiple Ingress/Egress points can be established into the enterprise, thus avoiding the single point of failure issue.

Further, IP-PBX manufacturers are beginning to see this problem, and as a result, are creating resilient PBXs whereby two IP-PBXs can be deployed in the enterprise to share telephone operations across the distributed enterprise.

But Even without such dual/resilient PBX products, one could simply configure station phones to work with two IP-PBXs in the enterprise, and then have that station information built into both IP-PBXs databases. That way, the phone is now served from two geographically dispersed switches when disaster strikes.

Figure 3 explains a redesigned basic network. In this architecture, a second ingress/egress point to the enterprise network is added, and also a second IP-PBX. The single point of failure in Figures 1 and 2 are eliminated.  This architecture radically reduces costs for voice communications and full voice redundancy on the customer premises is realized.

Now, let’s look at the same basic network redesigned in figure 3. In this architecture, we’ve added a second ingress/egress point to the enterprise network, and also added a second IP-PBX to act as a peer to the other system. In doing, so we’ve gotten rid of our single point of failure PBX but we’ve also done much more.

Since the ITSP can load share all traffic to the two or more ingress/egress points, we now have created a resilient enterprise voice network that cannot fail as the result of a single element being knocked out or a single interconnect to the carrier being disrupted. With this architecture, we radically reduce the costs for enterprise voice communications, and yet at the same time, realized full voice redundancy on the customer premises.

Here are a summary of some of the advantages of this architecture and technology.

  • Reduced CPE equipment - By having main switching facilities deployed in two locations, and only secondary IP-PBX’s or VoIP phones in the other locations, CPE equipment costs and the support costs that go along with them have become radically reduced.
  • One Network - With the TDM network in the enterprise eliminated and the IP network and staff taking over telephony functions, more significant savings are
    realized through economies of scale. Furthermore, simplicity and better reliability are also gained by extending IP technology into new applications, namely VoIP
    telephony.
  • True Global Number Portability - Ingress DID functionality is now freed from the TDM rate center, offering the VoIP network designer possibilities that
    previously were impossible (realizations of the Resilient VoIP network). See“Cyclix’s Global Number Portability” for a more in-depth discussion of this
    subject.
  • Resilient Services - Moving beyond TDM limitations, we now move into an era where voice can become truly redundant at the CPE edge as opposed to just the
    PSTN core.

Summary
It is curious to note that VoIP technology first started entering the enterprise on the station side. VoIP phones were implemented with the IP-PBX, with TDM trunks provided by the LEC and/or the long distance telephone service carrier. Now, with SIP trunking becoming mainstream thanks to ITSPs like Cyclix, the TDM trunk-side barrier is falling. We see captive expenses and limitations associated with the incumbent LEC and LD providers going away.

More important than saving costs, however, is the removal of TDM circuits on the trunkside of the enterprise, and replacement with VoIP. That allows true resilient survivable voice services to be realized for the enterprise. That is priceless.

< Page 1