The introduction of VoIP, as an alternative to legacy premise-based PBXs, has led to more types of network redundancy. While the premises-based PBXs have some contingency, they still have many points of failure. Offices that use these legacy systems are at the mercy of copper wires and no backups. Cytracom’s platform overcomes these problems while also improving upon the standard VoIP model as well. We do this with a new strategy called Active Geo-Redundancy.
The old way of doing things: Passive Geo-Redundancy
Before we review active geo-redundancy, let’s take a look at a commonly used architecture. Most hosted voice providers use a “Passive Geo-Redundancy” architecture to ensure redundancy for their network. This was the initial strategy Cytracom used when we launched our product over 11 years ago. What does that mean? It means that all traffic is pointed to a single data center or environment, and if there is some type of disruption, the phones fail over to the secondary environment. The problem with this strategy is that all the stars have to align for a seamless switchover to happen. You need an ideal outage (is there even such a thing?) and then the team responsible for executing the failover playbook needs to do so with perfect precision. The illustration below shows a typical passive geo-redundant communications network.
The reality is, it is very hard to simulate complex outages and even harder to train for them in mission critical environments.
The new way for today’s business communications: Cytracom’s Active Geo-Redundancy
Over two years ago, we started down a path of rethinking this passive geo-redundancy strategy and concluded the only answer was to go back to the drawing board and totally re-write the playbook. Enter the concept of an “Active Geo-Redundancy” architecture that requires no human intervention and no changes to cascade down to the users’ endpoint or network in the case of a disruption.
What is an Active Geo-Redundant network? It is when every client and users’ desk phone is connected simultaneously to two independent data centers or voice Points of Presence (POPs). This new paradigm means that customer phones are always actively connected to two live data centers. If there is a disruption with the primary data center, the end-user can re-originate or receive calls immediately from the secondary data center. There is no long failover process or manual intervention needed. See Cytracom’s Active Geo-Redundant architecture below.
To achieve this, it became clear very quickly that there was nothing off the shelf that was going to satisfy these requirements. We were going to have to build everything from scratch. These conversations led us to build the second version of our platform. Some of the major challenges we have had to overcome in the past two years to deliver this unique approach include:
1) Decoupling the management and orchestration layer of the platform from the voice processing. This was completed early this year. All of our web and API services have been migrated to a dedicated environment abstracted away from our voice processing centers.
2) Ensuring our desk phone would connect to multiple data centers at the same time. There have always been options to use DNS failover but that is not truly active-active. We wanted devices actively maintaining a registration to multiple redundant data centers.
3) The hardest part was ensuring that this new highly-distributed environment stays in sync with all the state changes that occur in a real-time communications environment, while at the same time having no dependency on any other region or facility to process calls.
We are very excited that our Active Geo-Redundancy Solution is now being rolled out to all clients and is on track to be completed by the end of the year.
What do you need to do to prepare?
These changes will automatically happen for all of your customers. However, as a best practice, you can double-check to make sure your firmware and network policies are up-to-date. You can find that information in our help center here.
Also, you can access Secure to see if you or your customers are currently in Active Geo-redundant mode. See the image below, where the connection status shows Geo Redundancy is active on this particular D2 device.
Our focus has always been on building solutions to make it simple for MSPs to meet all of their customers’ communications needs with one unified platform. We are excited to roll out more services like Active Geo-redundancy to provide a seamless communications experience and enable our partners to sell confidently knowing we continue to improve our reliability.
Contact your account manager or become a partner to learn more.