IPTV
Overview
As traditional telephony revenues decline [1],
service providers are increasingly
offering triple play bundles, comprising telephony, broadband internet and TV –
where the TV component consists of a mix of on-demand and broadcast IPTV. Examples
in the US include AT&T's U-Verse service and Verizon's FIOS, while elsewhere
France Telecom and China Telecom offer comparable services.
However IPTV (and particularly broadcast IPTV) is a challenging application for
service providers. Consumers are used to high quality and are intolerant of video
artifacts – which implies a packet loss rate of less than 10-6.
Similarly consumers expect high availability – which implies high levels of
redundancy and restoration in the network. Finally, consumers expect that the
service is responsive and interactive – with one measure being channel change times
of a second or less. All of this when a popular program or event may be broadcast
to literally millions of subscribers.
These characteristics create extreme challenges in core networks, and to some
extent in metro networks, where the infrastructure is a shared resource, and there are
many competing capacity demands from other applications.
Point-to-multipoint (p2mp) MPLS is one solution to the channel delivery
requirement. Quality of Service (QoS) and Traffic Engineering (TE) are built-in,
allowing service providers to meet the quality requirements, while optimizing use of
network infrastructure. Similarly, MPLS has well-proven schemes for high availability
using protection and restoration. Finally, p2mp MPLS scales well, particularly with
extensions such as LSP hierarchy with segmentation.
One approach is to use IP multicast to carry IPTV traffic in portions of the network
that are under the service providers effective control – such as in the Super Head End
or in the "last mile." Then use multicast VPN services over p2mp MPLS
transports in the core and metro networks, with a large subset of channels carried all
the way down to an intelligent edge device – so that they are readily available locally
to subscribers.
PIM and IGMP/MLD are excellent solutions to the responsiveness (and particularly
channel change) requirement. IGMP/MLD channel change requests from subscriber set top
boxes (STBs) are monitored ("snooped") at the edge devices, so that channel changes are
made locally – and in a timely fashion. The edge device also implements IGMP/MLD Proxy
function whereby channel requests are consolidated and forwarded via PIM to reduce the
load on the upstream network.
The diagram below shows a typical network delivering IPTV to triple-play
subscribers, and illustrates the approach described above.
Metaswitch has vast experience in supplying MPLS and Multicast IP Routing network
software products to communications equipment manufacturers, helping them build edge,
metro and core devices that help service providers deploy IPTV.
Features and Benefits
The Metaswitch solution for communications equipment manufacturers building
IPTV capable backhaul devices offers the following features and benefits.
- Portable and scalable implementations of the MPLS, IP Routing – Multicast, and IP Routing – Unicast standards
- Rich support for pseudowires, layer 2 and layer 3 VPNs, QoS and Traffic Engineering
- Full protection switching and fast restoration – delivering resilience within the network
- Built-in high availability and software upgrade – delivering resilience within a network element
- Engineering of the very highest quality.
Solution Elements
The Metaswitch solution for communications equipment manufacturers
building Carrier Ethernet devices is based on the following.
- DC-MPLS – including DC-RSVP and DC-LDP
- DC-IP Routing - Multicast
– including DC-PIM
and DC-IGMP/MLD.
- DC-IP Routing - Unicast –
including one or more of DC-BGP, DC-OSPF,
DC-ISIS,
DC-RIP,
DC-CSPF
and DC-RTM.
- DC-MPLS and DC-IP Routing protocol
support for (amongst others) pseudowires, layer 2 and layer 3 VPNs, QoS, traffic
engineering, end-to-end protection switching and fast restoration.
- N-BASE,
Metaswitch’s portable operating environment for network software products
- High
Availability Framework (HAF), Metaswitch’s architecture for delivering
fault tolerance and reduced downtime across the range of network software products
- Metaswitch engineering
services, including professional services, training, and support
directly from the Metaswitch engineering organization.
[1] ITU ICT Statistics 2007