The right marching order in core network virtualisation projects can save considerable time and money.
In 5G virtual core is de facto. With 2G, 3G and 4G core network virtualisation, companies seek synergies and cost-savings. For example, recently one of our operator customers announced that they will reduce the physical locations of their signalling network elements from 29 to 6 with virtualisation.
When planning the execution of core network virtualisation, signaling control plays a significant role. Why is this? Because it can, in fact, save you hundreds of thousands of euros.
Signaling control eases network operations and maintenance when traffic amounts and the number of network elements increase. Signaling control becomes especially critical in 5G, as the number of connected devices will increase dramatically, leading to massive growth in signaling traffic as well. Simply put, centralised signaling control is there to minimise maintenance costs, enhance network scalability and improve network security. It provides better steering and control of traffic and optimises capacity usage.
When executing core network virtualisation projects, operators may decide to start with the architecture directly connecting core functionalities that they need to run the business, such as Mobile Switching Subsystem (MSS), Home Location Registry (HLR) and Mobility Management Entity (MME). Quite logical, but it would be best to think again. Very soon the set-up will look something like the illustration below.
Picture 1. Ending up with a messy architecture in core network virtualisation
The above path will drastically increase the complexity, causing higher deployment and maintenance costs, especially when trying to scale up the network. Also, the integration of different vendors, Virtual Network Functions (VNFs), will become complicated because there is a need to cope with multiple different signaling dialects on a continuous base. Furthermore, the complexity will make the network more vulnerable to major network outage and long remedy lead time.
My hot tip to core network virtualisation projects is to start with signaling control. As signaling control is most likely needed anyway at some point, why not start realising its benefits all the way from the beginning?
I sometimes hear concerns and claims that executing signaling control in core network virtualisation is complex or difficult. In my opinion, it only indicates that it is executed in the wrong order, or without a unified approach.
If the starting point is the situation presented above (Picture 1), implementing a “mesh” approach will cause a lot of extra work. All Virtual Network Functions need to be reconfigured again. The operator ends up paying for double work.
However, if you start with signaling control, it is easy to:
Picture 2. Tieto Evolved Signaling Controller helps to keep up maintainability of core network virtualisation
When choosing a signaling control product which supports all network types, operational saving potential is significant. Instead of having multiple separate platforms to maintain e.g. EIR, MNP, signalling FWs, thanks to Tieto Evolved Signaling Controller, there is only one to take care of.