When considering the migration to IP video, Patrick Wright-Riley, director/Strategy and Business Development at Motorola Mobility, believes it’s really more about integration.
“QAM is a big investment,” he noted at this week’s SCTE Cable-Tec Expo in Atlanta, so the question is how to integrate IP delivery to multiple screens on top of the existing set-top population, especially as set tops are expected to stay in the network for a long time.
“One MSO we talked to expects to still have millions of set tops deployed in 2025,” reported Wright-Riley. “The QAM set tops deployed can make you money for a long time to come. There’s no reason to get rid of them.”
Once the migration begins, Wright-Riley expects it to proceed in stages. One possible scenario looks like this:
- Deploy SDV to reclaim bandwidth used by long-tail content and MPEG-4 to add HD services efficiently.
- Deploy DOCSIS 3.0 to enable channel bonding to serve businesses.
- Roll out digital television adapters (DTAs) to recover expensive analog bandwidth and cost-effectively serve basic subscribers.
- Deploy start nDVR and Start Over to generate revenues, match competition and make all set tops DVR-capable.
- Deploy VOD in the cloud to decrease complexity and cost to deploy VOD.
- Convert second-tier broadcast to unicast, which completes the migration to IP and adds capacity for unicast services.
- Adjust bandwidth allocation as needed to recover the last of analog bandwidth and to adjust IP services to meet customer’s needs.
Cox Moves Video To Backbone
Cox Communications began its migration of national broadcast video to an IP backbone in 2007, explained Mark Pelligre, senior manager/Transport Networks. Dubbed V2B 1.0 (Video to the Backbone), that initial effort established a Cox-owned national DWDM backbone and centralized acquisition of all national video content at two master headends – one on each coast.
V2B 1.0 brought Cox many benefits. It eliminated duplicated efforts in multiple markets, improved the rollout of new national-broadcast content like high-definition video, reduced expenses for acquiring content and maximized video quality by delivering consistent quality across all markets. It also had several shortcomings, Pelligre admitted.
Scaling the V2B 1.0 architecture beyond 6 Gbps was difficult on 10 Gbps links; it was taxing on the Layer 3 backbone capacity. It also required significant engineering and planning.
“Video is the highest QoS, so everything had to be engineered around that. It’s a big elephant that needs to be accommodated,” said Pelligre. Troubleshooting also was challenging, as failover behavior was nonintuitive.
To handle those challenges, Cox launched its V2B 2.0 architecture in 2010. With V2B 2.0, Layer 1 multicast replaces Layer 3 multicast. Dedicated wavelengths carry all linear broadcast video traffic on the network. The solution uses two counter-propagating wavelengths, one for the East-Coast headend and one for the West. The result is four total feeds available to each market, which Cox says greatly improves reliability.
The benefits of V2B 2.0 are significant, said Pelligre. Costs are reduced because broadcast video traffic is removed from the routers. Scalability and service velocity are improved, as new channels can be added just once at the master headend and immediately available at all markets. Such network modifications as adding or deleting markets, drop points or new fiber links also are easily enabled with simple cross connects. Because failover behavior is predictable and intuitive, Cox can also better manage troubles, Pelligre concluded.