Cable MSOs and edge compute: what are executives thinking?
Cable operators are deploying Distributed Access Architectures (DAA) and extending fiber closer to the customer. Does this mean cable operators are well positioned for edge compute (moving compute and storage closer to the edge)? Are cable operators actually moving to edge compute – if so, why? Where is ‘the edge’ for cable operators – the headend, the primary or secondary hub sites, other locations?
These are some of the questions Broadband Success Partners discussed with network and business services executives at cable operators of all sizes in North America.
We had an opportunity to get some preliminary insights from David Strauss at Broadband Success Partners and to hear from Fernando Villarruel, Chief MSO Architect at Ciena regarding additional considerations as cable operators move toward edge compute. The final research results will be presented on a webinar on April 23, in the meantime here is your sneak peek at three fundamental questions for edge computing and the cable industry.
1. How do cable operator executives view edge compute?
David: There’s no single definition. Cable operators are deploying edge computing (or expect to) in one of three ways. According to 43% of those interviewed, transforming headends and hub sites into mini data centers, or Headend Re-architected as a Data Center (HERD) is how they describe their edge compute initiatives. Distributing compute and virtualization via Flexible MAC Architecture (FMA) was cited by 33% of the executives. The remaining 24% view edge compute as building new edge sites with compute and storage closer to end customers.
The reason for these varying views is likely due, in large part, to the individual’s view of edge computing uses cases. For example, if their primary applications are less latency sensitive such as video caching or SD-WAN, they skew toward a less distributed compute architecture. In contrast, those thinking in terms of AR/VR, gaming or autonomous vehicles with little to no tolerance for latency gravitate towards another configuration.
Fernando: With cable operators using virtualization and cloud environments to deliver applications and networking functions, they will need to view edge compute by thinking about how they can best distribute software throughout the network. This can be for optimized throughput and new services, operational efficiencies, or improving the quality of experience. Edge compute provides a practical framework on the journey to next generation networks.
2. Where is ‘the edge’ for cable operators?
David: Many of the executives we spoke with said their goal is to get as close to the edge of the network as possible. Headends at 33% and hub sites at 29% were the top responses as the most likely locations for ‘the edge’ for cable operators.
An interesting split surfaces when you view the results by tier-1 and tier 2-3 cable MSOs. Tier-1 MSOs are more likely to place edge computing hardware in hub sites rather than headends. The opposite holds true for the tier 2-3 MSOs. This is due, in part, to the cost of scaling up and out to place edge devices in many hub sites versus fewer headends. This investment needs to be considered relative to the allowable latency.
The location of the edge is also dependent on where cable operators have space and power available – headend facilities may be more viable in this regard compared to hub sites. The density of the network could also be a variable – in rural systems the edge compute and storage could be in headends, but in hub sites in higher density suburban systems. Several of the executives indicated they are starting with less costly headend deployments (fewer) and will then extend the edge to hub sites at a later date.
Fernando: The research reveals that the location of ‘the edge’ will vary by MSO and on the drivers for moving to edge compute. What needs to be considered is how the system will flow end-to-end. With edge compute there are several fundamental items that will have to be addressed:
- Awareness of end-to-end connectivity from the headend to the edge
- State aware command and control for the applications and network functions
- AI, analytics and policy management for dynamic network management
3. What are the drivers for cable operators to move to edge compute?
David: Over half of the executives noted either improved customer experience or the enablement of new revenue streams as the most important drivers for edge computing in their company. Improved customer experience was the top driver for 29% of the executives - lower latency for gaming and video optimization were use cases cited.
Enablement of new revenue streams was the top driver for 24% - new revenue streams for both consumer and business customers.
Fernando: Improving the customer experience and enabling new revenue streams with edge compute will require cable operators to have a highly programmable infrastructure paired with robust management, assurance and orchestration, where network telemetry creates data rich in analytics options for lifecycle management. It is an exciting time for the cable industry as we embark on this journey enabled by the transition to DAA.