A distributed network architecture requires distributed OSS/BSS functionality
As it considers the move from 5G Non-standalone to 5G Standalone, alongside deployment of mobile edge computing (MEC) infrastructure, Three UK is mindful of the concurrent need to evolve OSS/BSS functionality. Speaking at the recent Telco Cloud and Edge Forum (available on-demand), Three’s Senior Manager of OSS Strategy and Architecture Ankush Saikia linked together the technology and organizational moves necessary to manage increasingly complex network and monetize advanced connectivity/compute-enabled applications.
In terms of 5G Standalone, he characterized the transition as “one of the key drivers for the telco cloud at the moment.” He said OSS would carry-on with its core function of network management and monitoring “but it’ll play a more important role in the 5G SA world.” In a situation where an operator has moved to a cloud-native architecture and operating model and linked up its network with distributed computing nodes, “What it means for OSS is that our direct collection points have increased. Instead of collecting data from these core locations, we now have to collect data from every other edge location.” This equates to added complexity around data collection and sheer data volume. “OSS/BSS platforms have to process data at speed to provide a near-real-time view to the operations people.”
This points to another type of evolution around the people and processes that use operational support systems. “The continuous cloud brings us the opportunity to bring automation via DevOps, CI/CD pipeline, and OSS will be the center to that,” Saikia said. “The continuous cloud also means it’s very complex to manage, especially when you consider that there will be a period of time when operators will have virtualized cloud which will be hosting the 4G, and the continuous cloud which will be hosting the 5G. So managing this kind of hybrid architecture or hybrid network will need a lot of intelligence which we cannot build only manually.”
Saikia also discussed how mobile edge computing, as referenced above in commentary around managing and monitoring a network with more data-generating endpoints, requires an architectural overhaul of OSS/BSS. “That architecture needs to be distributed along with the network,” he said. “And we should be focusing on processing the data at the edge locations.”
He referenced ambitions around network slicing and ultra low latency use cases as heavily dependent on a robust OSS; to sell a slice or an ultra low latency service against a service-level agreement, operators need a new level of visibility and control. “I think one thing we need to bear in mind is the scalability,” Saikia said. “This traditional box system is not going to work. Because of the amount of traffic, of volume, it’s expected to increase exponentially along with 5G NR. So I think scalability is going to be on the key factors in selecting an OSS/BSS vendor.”
In summary, Saikia cautioned to not repeat the mistakes of the past—the move from 3G to 4G, for instance, when OSS/BSS was “an afterthought. That has created a lot of design gaps, lot of capability gaps…I strongly suggest that you put OSS/BSS at the forefront…along with the network so that we do not create any gaps upfront, and we reach the nirvana state of a single pane of glass very quickly in managing a telco cloud.”
