Each subsequent step northbound in the conceptual diagram below does the same thing:
- It abstracts (vertical) – it only performs a sub-set of the lower layer’s functionality
- It connects (horizontal) – it performs the task of connecting and managing a larger number of network elements than the lower layer
So an OSS, will never (almost never?) replace the functionality of the NMS or EMS. But in return it should increase the cross-domain connectivity.
However, when abstracting, you’re effectively minimising or making a determination on what to keep and what to discard.
Does your OSS keep track of what it’s discarding? For example, does it have the ability to reconcile all alarms collected at each NE against the list of alarms that bubble up through other layers to reach the OSS? It may have caught all alarms when the OSS was first commissioned, but there have been various upgrades in hardware and software since then that may have introduced new capabilities that the OSS doesn’t know to look for. This exacerbates in highly transient networks, such as those being introduced by network virtualisation.
Do you know what you’re missing?