The problem with today’s OSS infrastructure is that it’s based on a functional silo model.
This functional silo model started at a time when:
- There were less technology network domains
- Those domains were much more standardized
- Technology changed much more slowly
- Computing technology did not allow multi-function systems that could adequately scale.
This leads to the fundamental issue with the current OSS model…the lack of abstraction.
Every OSS system (Inventory, Fault, Provisioning, and Performance) needs to understand the details of all the different network domains, which results in a lack of data integrity across systems.
This images helps to understand the complexity.
Each of the OSS systems has to do its job across all domains.
For example, the Inventory system needs to know about:
- TDM domain: The boxes, cards, ports, and fibers that make up this domain. The logical entities of this domain like existing channels, whether it is BLSR or UPSR, etc.
- ATM domain: The Inventory system needs to know about the EVCs and COS.
- Ethernet domain: It needs to know about boxes and cards, but also VLAN IDs, policers, shapers, reverse path learning, and protection methods.
- IP MPLS domain: The Inventory system needs to know about LSPs, the control planes that set these up, LLDP, VPLS instances, and so on.
This is the same for every OSS system in the network – it is phenomenally complex.
Then, when you add SDN and NFV elements into these existing networks, this further complicates the OSS infrastructure.
The lack of abstraction in the OSS infrastructure leads to serious problems for network operations personnel. Any change in any domain means manually having to having to update the software of each OSS to support this change. A change in any OSS system means ensuring the change is supported across all domains. There is no hiding the complexity within the layers of the network and the result is a vertical functional silo model, as opposed to a horizontal layer that does the abstraction.
So how do network operators overcome this fundamental problem with their existing OSS infrastructure? How can they abstract this complexity and prepare themselves to begin introducing SDN and NFV elements without further adding to this complexity?
The answer: Service Orchestration.
To learn more about the importance Service Orchestration pays in unleashing the full potential of SDN and NFV, download the white paper, “Service Orchestration for Next-Generation SDN & NFV Networks”.