Cisco IPTV Traffic Forecast

Cisco has released its latest IPTV traffic forecast that covers 2008 to 2013. It finds that video traffic will be of increasing importance and will account for 80 percent of consumer Internet traffic by 2013. It states that Internet traffic and consumer IP traffic will account for more than 70 percent of total traffic by 2013. This means that more than half the traffic on the network will be video traffic at that time.

This reports shows IP traffic growing at a compound annual growth rate of 40 percent per year over this period. This is a bit below the 50 percent per year that I have been assuming. My assumptions are that video on demand to the TV will grow faster than Cisco's forecast. I have seen very high rates of video on demand utilization where network PVR services are available.

It is clear that IP traffic will continue to grow at a high rate and that video will come to dominate these networks. This will require the architecture of these networks to change to optimize the transport of video traffic.

Broadband Providers "Not So Fair Use" Violates Net Neutrality

A couple of articles have been published recently describing how broadband providers are managing their traffic to their advantage and to the detriment of specific services. The BBC has published an article describing how BT throttles the bandwidth on its broadband service and degrades the quality of the BBC iPlayer at certain times of the day. The second is a Light Reading article describing the results of femto cell testing by Epitiro that shows that traffic management by broadand providers will negatively affect the performance of femto cells during peak traffic periods.

The BBC has identified a clause in BT's fair use policy that permits it to reduce the bandwidth on its broadband service to below 1 Mbps, which degrades the quality of the BBC iPlayer over the top IPTV service. The article cites an example of a user who has been affected by this policy.

The traffic analysis performed by Epitiro shows that some broadband providers are throttling IPSEC traffic so that it provides worse performance that best efforts Internet traffic at the same time.

These are the kind of situations that those promoting the Net Neutrality agenda are concerned about. In both cases the broadband providers are throttling the traffic of competitors. The BBC iPlayer competes with BT's BT Vision IPTV service. Femto cells will compete with the broadband provider's VoIP and POTS services. The broadband providers are building the case for Net Neutrality when they do this and will produce a regulatory reaction that they will not like at all.

Cox Cable's Traffic Management Strategy

Light Reading published an article describing Cox Cable's strategies for traffic management. Cox's pilot approach divides traffic into "time-sensitive" (e.g., Web pages, voice calls, streaming videos, games), and "non-time-sensitive" (e.g., file uploads, peer-to-peer, and Usenet) categories. The system will delay the upstream, non-time-sensitive traffic when network congestion is detected. All traffic returns to normal when the congestion abates.

This is the right approach - classifying traffic by its response time requirements and delaying lower priority traffic during periods of congestion. This is a much more reasonable approach than the monthly usage caps that other operators are implementing.

Cisco Releases Lastest Traffic Study

Cisco released its latest Visual Networking Index study that updates its global data traffic forecast. Key findings include:

• Global mobile traffic will increase 66-fold between 2008 and 2013 based on a 4G mobile Internet.
• Nearly 64 percent of the world's mobile traffic will be video by 2013.
• 4G mobile broadband devices will constitute more than 80 percent of global mobile traffic by 2013.
• Latin America will have the strongest mobile growth at 166 percent CAGR, followed by the Asia-Pacific region at 146 percent.
• Asia-Pacific will account for one-third of all mobile data traffic by 2013.

The forecast that video will account for nearly two-thirds of mobile traffic by 2013 is interesting. My report Investing in Mobile TV took a close look at the business case for mobile TV and found that it will be difficult for carriers to make money on such a service. It generates too much traffic to produce much profit. It will be interesting to see how this works out.

Inauguration Strains the Internet

The inauguration of President Barak Obama on January 20 put a significant strain on the Internet. There were reports of difficulties connecting to the broadcast.

Arbor Networks published and article that shows the traffic peak that occurred. Backbone traffic jumped by as much as 40 percent over a normal Tuesday.

This peak occurred with most people using lower speed streams intended for PCs. Think of what would have happened if this content were be delivered to HDTVs at 1080p resolution!

Beefing up the Internet to handle this kind of demand but of HD streams is the challenge that the major ISPs have to face to make Internet TV more than a side show. This experience showed that the Content Delivery Networks were effective in managing backbone traffic, where they were used. This leaves the metro networks. My report Telecom 2020: Transformation Strategies suggests that video needs to be kept at the optical layer in the metro networks to handle this kind of onslaught. It will be interesting to see how this plays out.

First 100 Gbps Network for ESnet

Light Reading has published and article describing the 100 Gbps network that ESnet and Internet2 plan to deploy by 2010. This network will use equipment from Infinera and Juniper Networks over Level3's network to combine ten 10 Gbps links into single 100 Mbps logical links. This is in advance of when 100 Gbps Ethernet will be broadly deployed.

ESnet has found that it is currently supporting single flows of 10 Gbps and needs the 100 Gbps links to be able to handle them along with other traffic.

The U.S. Department of Energy has long been a leader in super computing and high speed communications. This is an example of the kind of application that will drive the speeds of networks up to and beyond 100 Gbps during the next decade.

Telco IPTV Network Architecture

Both Alcatel-Lucent and Cisco have introduced edge routers that facilitate Telco IPTV networks by supporting fast channel change, packet retransmission, and ad insertion using a large flash memory in the router. Both companies are talking about having enough memory in these routers to support video on demand caching. That is, the most popular video on demand assets would be stored on the routers which would eliminate the network traffic required to provide them from a centralized server.

I think that these developments are a real step forward for IPTV networks. I think that their ability to support fast channel change and packet retransmission are fundamental improvements. The ad insertion and video on demand capabilities will provide significant improvements, up to a point.

The fundamental issue will be the amount of storage provided by the router. In a personalized ad insertion system, this will work fine as long as the number of personalized ads does not become too large. The issue with video on demand will be the hit rate, that is the number of video on demand requests that are served by the content on the router vs. the centralized router.

My own thinking on this issue has evolved. I did a report for MRG in the scaling of IPTV networks three years ago that identified that the amount of traffic required to support high levels of on demand traffic will be a major issue. At that time, I felt that the caching approach that Alcatel-Lucent and Cisco are talking about would be the answer.

I looked at this issue again last year in the report Networking Strategies for TelcoTV Services. With further analysis, I realized that the traffic to fill the distributed video on demand caches would be enough by itself to swam metro networks. My conclusion was that the IPTV traffic had to be removed from the IP/Ethernet networks and move down to the optical level.

My recent report Telecom 2020: Transformation Strategies looked at the issue again and took into account the consolidation of fiber access systems into large centers that would serve very large numbers of subscribers. This architecture would mean that there would be enough video on demand traffic to one of these major fiber access centers to fill a 10 Gbps or even a 100 Gbps optical link; therefor, there is no benefit for bringing this traffic up to the packet layer. It will be much more economical just to pass directly to the fiber access center over a direct optical link.

There are too philosophical points that drive my thinking. The first is that as IPTV moves from multicast to on demand and unicast, the TV traffic will dominate the network. It just makes sense to optimize the network for the dominant traffic type.

The second point is that it is very difficult to understand how a packet network can provide an acceptable SLA or QoS when 99 percent of the traffic on the network is deterministic TV traffic. On the other hand, it is easy to see how this will work over direct optical links.

There is a real question of timing here. For many carriers, the cached approach introduced by Alcatel-Lucent and Cisco will be a big help over the next three to five years. However, I do think that these carriers will be thinking seriously about a direct optical approach after that.

Cisco Introduces 100 Gbps Ethernet Edge Router

Cisco introduced its Aggregation Services Router (ASR) 9000 that will provide significant increases in performance over its work horse 7600 routers with a total of 6.4 terabits per second of capacity. The ASR 9000 is designed around 100 Gbps Ethernet interfaces.

The ASR 9000 incorporates the Cisco Advanced Video Services Module (AVSM) with 4 terabyts of flash RAM storage that offers content caching, ad insertion, fast channel change and error correction of video streams. Fast channel change and onboard error correction for both unicast and multicast video traffic helps ensure that errors can be detected by any set-top box and retransmitted within milliseconds.

This is the kind of improvement in performance that will be required as we move through the next decade as 100 GBps Ethernet links become commonplace.

The AVSM capabilities of the AVSM provide similar capabilities to Alcatel-Lucent's TPSDA 2.0 approach that provides similar capabilities. Both approach will require support from the IPTV middleware to get their full effect.

The 4 TB of storage will be enough to store about 1,000 hours of HD content. This will be enough for the most popular video on demand assets but not for the long tail of for an extensive NPVR service. Cisco will have to add a lot more storage to really offload video traffic from the network as it is currently claiming. I will write a column that will discuss this issue in more detail in a couple of days. I just don't think that a distributed approach will hold up over the long term.

How Much Will Network Traffic Grow?

Probably the biggest issue in planning for a network for 2020 will be the amount of trafficthat it will have to support. Network traffic has been growing by about 50 percent per year over the last 10 years. At this rate, the amount of traffic will be 100 times what it is today.

The question is how long will this party go on? It seems reasonable that this torrid rate of growth will have to moderate at some point of time. Any decrease in this rate of growth will certainly make life easier for the carriers. They would be able to reduce their investment in their networks.

It is clear that wireless networks are just now starting to see the impact of the growth of data services on their networks. The iPhone and similar devices are bringing a new set of Internet users to mobile networks. This is only the start. During the next decade most mobile users will be making extensive use of data services.

This means that the growth of data usage on mobile networks will probably grow faster than 50 percent per year over the next several years and grow at a rate similar to wireline networks after that.

It is hard to see what will continue drive growth on wireline networks at such high rates. I believe that it will be video and TV content. TV content will particularly important based on the increasing popularity of HD and 1080p.

Over the last 15 years there have been a steady stream of new applications that generate ever increasing amounts of traffic. I am not going to bet that this innovation has stopped. I think that anybody thinking about networks for 2020 must assume a factor 100 increase in traffic.