For the first of our interviews with keynote speakers attending the Broadband World Forum event, taking place in Paris this October, we speak to Stefan Dyckerhoff, Executive Vice President and General Manager of the Infrastructure Products Group at Juniper Networks.


What changes have you observed in the broadband industry over the past 12 months or so?

The overarching trend that is painfully clear to service providers is rapidly increasing network traffic and bandwidth consumption, with a concurrent erosion of prices. A key driver of the traffic increase is video, but it is also exacerbated by the proliferation of mobile devices and the popularity of mobile Internet. To handle the mobility side of the equation, Juniper is investing significantly in Project Falcon, which optimises mobile data traffic to transform not only the economics, but the experience of mobility as well.
 
Other key trends are virtualisation and cloud computing, which are really about the need to consolidate and simplify the data centre to ensure efficiency in delivery of content and new services while scale requirements continue to grow. Our solution to this need is Project Stratus, which flattens networks, with a single fabric uniting Ethernet, Fibre Channel, and Infiniband networks. This vastly simplifies the operation and reduces the cost of running a data centre network.

 
Do you think there will be sufficient demand to justify eventually moving to all-fibre networks?
 
There is no doubt that all-fibre networks make sense in many parts of the world, including most metropolitan centres. The demand is there, and growing, evidenced by increasing traffic. Almost 25% of the world’s population will be Internet users this year. For most of that subset of the population, we can expect that fibre networks would be a boon now, and far into the future.
 
However, it really comes down to a question of economics. For service providers, it just does not make economic sense to build out fibre networks in some areas, despite the superior experience that would be enabled. I think it is likely that multi-tiered demand will continue to drive multi-tiered broadband solutions, for the foreseeable future.
 
The global financial crisis may also have a silver lining for our industry, as governments around the world acknowledge that broadband plays a key role in economic development, positive social change and national infrastructure and, as a result, invest more in their broadband infrastructure.
 
 
What areas do you think particularly need improving from a technology point of view?
 
The single greatest challenge for the broadband industry is ensuring the long-term economic viability of the broadband networks upon which the world is becoming more and more reliant.
 
For service providers, the infrastructure investments required to accommodate explosive increases in traffic demands are growing steadily. At the same time, increases in revenue from operating the networks are slowing. Based on this trajectory, we predict that around 2012 to 2015, the Internet business model may “break,” as necessary network upgrades and maintenance become economically unviable. Could this be what the Mayan calendar really predicts? (kidding) But seriously, imagine the impact on the global economy and infrastructure if this actually occurs!
 
The only way around this scenario is to overhaul our approach to network planning and architecture. The incremental “stacking” of additional hardware and functionality will only prolong the dying gasp of a legacy network architecture that is incapable of economically accommodating growing demands.
 
The first step in “future-proofing” the network is really a matter of designing flexibility and scalability that is capable of providing virtually limitless service. Because of its inherent flexibility and scalability, software is a critical component of the design process. By building multi-purpose hardware with functionality largely determined by the software, and building it intentionally for flexibility and scalability, we create a foundation that can be leveraged for a wide range of scenarios that may develop.
 
The second step is acknowledging that a single company cannot anticipate what applications will prove popular and profitable in the future, and recognising that innovation is occurring in garages around the world. Opening the network to an ecosystem of developers can enable unprecedented access to innovative applications and solutions with fast time-to-market that can also help ensure the future economic viability of the network.