Demand for bandwidth to the household continues to grow, and cable multiple system operators (MSOs) are racing to meet the need. As technology advances, however, traditional methods to expand capacity are being rendered either technically unfeasible or cost-prohibitive.
For example, the practice of segmenting nodes to reduce the size of service groups has become a popular way for many operators to increase bandwidth available to each subscriber -- in some cases, doubling or tripling the number of nodes in just two or three years to enable delivery of gigabit services. Yet as operators add networking equipment to support the increasing node counts, rack space and power-cooling capacities at hub sites are stretched to their limits.
Distributed Access Architecture (DAA) technologies such as Remote PHY address these constraints and reduce space and power requirements in the headend. Relying on commodity optical Ethernet technology, this approach promises affordability and flexibility. In fact, roughly 80% of MSOs have DAA deployment plans today, hoping to reduce total cost of ownership (TCO) and improve profit margins.
However, these new technologies present a number of deployment and operations challenges. How can MSOs manage these challenges, and what do they need to know as they plan their Remote PHY rollout?
Room to grow
As the name implies, DAA addresses space constraints at hub sites by distributing portions of the access architecture out to the field. Remote PHY essentially splits some of the hub-based functions and redistributes portions out into the fiber nodes. Some new distributed architectures promise to reduce space and power requirements by up to seven times.
After additional fiber has been built out into the field, closer to subscribers, MSOs will have a more flexible hybrid fiber coaxial (HFC) network that can be used for other improvements. These might include addressing bandwidth usage imbalances in a user group, supporting backhaul for the numerous cell sites that 5G mobile data requires or delivering Ethernet services to business customers.
Moreover, Remote PHY provides improved optical link performance, migrating analog optical links to multiple digital Ethernet links (typically 10G) by using existing technologies such as Dense Wavelength Division Multiplexing (DWDM) to enable maximum benefits from the latest DOCSIS protocols. This delivers the capability to transmit more bits/second/Hz to the network edge to boost capacity.
But it's not all good news: These architectures also expose field operations teams to added responsibilities and challenges they have yet to face, requiring changes in the processes and tools used to maintain peak efficiency and ensure quality of service (QoS).
Mind the gaps
For today's MSOs, every bit of capacity is important to maximize return on investment (ROI) and minimize TCO. To pass more packets and fully utilize the higher modulation orders available from DOCSIS 3.1, cable plants must be clean -- and testing is the only way to ensure a clean RF plant.
Current test and maintenance procedures rely on RF test points for hub-based monitoring gear, such as signal-leakage detection, return path monitoring and sweep. But the redistribution of the PHY layer capabilities out into the field means that RF is eliminated from the hub. To address this architectural shift, network technicians must adopt new, innovative approaches, such as virtual test solutions that leverage the Remote PHY Device (RPD) to handle the critical tasks of spectrum analysis, sweep functionality and leakage tagger capabilities.
Turning up Remote PHY nodes will also require the certification or validation of the fiber physical layer links feeding the new nodes. This ensures that newly activated DWDM wavelengths correctly route from hub to node through any optic combiner/splitter (MUX/deMUX), and that the DWDM channels supporting various Remote PHY nodes are present and demonstrate the correct optic power levels. This applies equally to newly laid fiber as well as existing analog fibers being migrated to digital; after all, those existing fibers may have been carrying light previously, but not at these new wavelengths.
Home in on the opportunities and challenges facing European cable operators. Join Light Reading for the Cable Next-Gen Europe event in London on November 6. All cable operators and other communications service providers get in free!
Complicating the job of maintaining the cable plant is the blurring of lines that results from creating a new network interface. By migrating network functions that were previously hub-based out into the field, traditional separation of functionalities is disrupted for existing MSO groups and processes. With DAA technologies, the split between headend and field is not always obvious, creating the potential for fingerpointing between workgroups or issues to fall between the cracks when troubleshooting.
Time for a change
One particularly tricky complication is that of timing. Network timing has always been critical for DOCSIS communications, but this has been relatively simple to maintain by colocating the Media Access Control (MAC) and PHY layers of the network stack in a common equipment set. The MAC sublayer in Layer 2 functions as a controller that defines how and when modems on a network transmit over shared spectrum without impeding each other, like a traffic cop. Now, with the MAC-PHY separation that occurs with Remote PHY architecture -- often by long distances and several network hops -- synchronization can become a real issue, creating the potential for service degradation and packet loss.
Stop. Now Go
Like a police officer directing traffic, the MAC sublayer in Layer 2 acts as a controller that says when and how modems on a network can transmit over shared spectrum, ensuring they don't impede each other.
To prevent QoS problems, network teams should become familiar with turn-up, maintenance and troubleshooting processes from the Ethernet domain, such as testing according to the IEEE 1588 Precision Timing Protocol (PTP) for packet-based transmission. Otherwise, field technicians could spend days troubleshooting upstream bit-error-rate issues resulting from PTP timing issues, but would never have a chance of properly diagnosing them, much less addressing the root cause.
Additionally, as new architectures are added to legacy networks, virtualization is sure to play an important role as well. Field instruments will need to evolve to interact with both physical and virtual test systems to minimize mean-time-to-repair (MTTR) of service issues and avoid subscriber churn.
Although new DAA architectures like Remote PHY may be operationally disruptive in the short term, the promise of more throughput to subscribers at a lower operational cost is hard to pass up. To ensure maximum ROI throughout migration and beyond, it's time for MSOs to deal with the idiosyncrasies of DAA technology at all layers of the stack.
— Jim Walsh, Solutions Marketing Manager, Viavi Solutions. Follow Viavi on Twitter @ViaviSolutions, on Facebook or LinkedIn.