TCO reduction is typically the leading driver for Open RAN. No doubt TCO reduction is important, but it only tells part of the story. Operators around the globe are looking to Open RAN for increased opportunities as well as cost savings. Operators can have both. Open RAN could deliver up to 30% TCO savings and help service providers access the $23.5 billion of revenue opportunities[1] expected from far edge services in a wide range of new enterprise use cases. Open RAN clearly has potential for operators who make the right investments in technology and skill sets now.
Moving to Open RAN is also expected to reduce time to deploy new services, improve scaling and elasticity, support new enterprise revenue and improve the business case for mobile edge services. The benefit of these 5 elements taken together could make TCO reduction seem small by comparison. In addition, there are benefits for the ecosystem such as, accelerating innovation, futureproofing networks, and diversifying the supply chain. These may be a bit harder to quantify, but it’s hard to argue they wouldn’t represent positive impact. The point is, TCO reduction is important, but when considered with the myriad other Open RAN benefits, it doesn't take much reduction to justify the move.
Quantifying the TCO
A recent study by Analysys Mason looked at three different Open RAN architecture options. Two examples (distributed and vCU centralized) represent what is being deployed today and one (vDU/vCU pooling) that will likely be more common as the edge gets built out. Those three scenarios were modelled for three different operator profiles including a Tier-1 operator in Western Europe, a medium-sized incumbent operator in a developed market and a Tier-1 operator in an emerging market. Within a 3-year horizon, TCO savings is expected among all architectures and profiles with a range between 13% to 30%. The savings increased up to 35% with a 6 year horizon.
Table stakes
TCO savings will be calculated across a variety of variables. While this study did not include every single variable it did include: RAN hardware and software, backhaul, supplier support and maintenance, additional FTE costs, power and rackspace costs. Beyond the assumptions and basic costs, the study found in order to be successful, Open RAN must include these three key pillars:
Highly performant and secure Kubernetes based O-cloud infrastructure that supports:
- Highly demanding processing requirements of the 5G RANs real-time functions as well as advanced technologies such as MIMO. Optimization of the number of server and power requirements in order to reduce costs to deploy and scale the distributed hardware footprint. Centralized and distributed RAN architecture.
- The execution of decomposed, cloud-native microservices on bare metal infrastructure
- Time-synchronization and heterogeneous architecture with multiple accelerators while retaining openness and programmability
- Cloud-native network functions (Open RAN and others) from multiple vendors as well as other enterprise and consumer service applications
- A high-level of security and reliability to host all network functions and applications with strict isolation between them.
Distributed orchestration and zero-touch automation
- Open RAN needs zero-touch automation and orchestration that spans the entire lifecycle, including Day 0 deployment (remote, automated software installation), Day 1 (preparation for operation, tests and validations) and Day 2+ ongoing live operations (self-healing, self-optimization, updates and upgrades with CI/CD)
A pre- validated ecosystem
- A pre-validated ecosystem of Open RAN vendors that can provide an open, standards-based full-featured stack to reduce integration costs and time to test/deploy.
Attempting to deploy Open RAN with something less than the above would result in far less TCO reductions and could potentially result in cost increases.
Looking to the future
Once the infrastructure is in place the expectation is that costs will continue to decrease, which underscores the importance of starting with the right components now. As Open RAN RU and antenna reference designs become standardized it will lead to commoditization, driving down the cost of these components. Chipset innovations could result in new server form factors with lower cost and better efficiency. Plus, next-gen CPUs are expected to reduce the number of physical cores required to run vDU and vCU cloud native functions, and the O-cloud platform, which would lead to fewer servers and/or more resources for other applications.
Conclusion
Open RAN offers opportunities to cut costs and realize new revenue when done correctly. Considering the many benefits, Open RAN is worth the move.
