29 November 2014

Softwarization: two scenarios towards 5G

In the IEEE SDN initiative we wish dealing with SDN-NFV adopting a “systemic” perspective.

In fact, I believe that Cloud (and Fog) Computing, SDN and NFV are just different expressions of the same systemic trend (impacting not only the Telco networks by also several other segments and Industries) which is called “Softwarization”. This trend has been/is triggered and steered mainly by ultra-broadband diffusion (i.e., high bandwidth and low latency connections) and the IT techno-economic drivers (e.g. increasing performance of chipsets, at continuously decreasing costs). This wave of innovation may have different levels of impact on the Telco infrastructures, depending on several variables, among which the speed and level of adoption of “Softwarization” by said industries, the foreseen models for business sustainability and the regulatory issues.

Personally I believe that these trends will potentially bring soon into reality (at least) two main scenarios, characterized highly different techno-economic perspectives:

·    Scenario 1: an evolutionary scenario, based on a smooth and gradual introduction of SDN (coupled with NFV) into the legacy Telco networks, e.g., starting from virtualizing network functionality; communities looking at this scenario are focusing on short-medium term challenges such as interoperability aspects of virtualized network functions with the systems of current infrastructures, integration of orchestration capabilities with the OSS (e.g., ETSI reference model), Standardization of relevant interfaces, etc.

·         Scenario 2: a disruptive scenario (where speed of adoption of innovation is much faster, so that it will jumps ahead any gradual evolution of current infrastructures as described in Scenario 1.). In this case “Softwarization” should be seen from a more “systemic” perspective, which is going far beyond the network and the Cloud, but it is reaching also terminals, smart phones, machines, smart things, robots, etc. The market will decide the standards the facto.

This second scenario is clearly disruptive. A main point  is the “lowering of the threshold” for other Players to enter the Telco market to offer services (as potential competitors).  As a matter of fact, this progressive “threshold lowering” is driven by chipsets technology advances and cost reductions (bringing to more and more powerful IT systems at lower and lower costs), and by the large availability of Open Source software solutions (examples are the open source operating systems for the Cloud and the network, respectively OpenStack and ON.OS, etc).

In this sense there is the risk a number of other Players (even with limited investments, much lower than in the past) will be able to provide services (with different business models) using low-cost software platforms of virtual resources, spanning from the terminal to the cloud (Opex-oriented model), and by using “lean” operational processes. This would mean that the gradual evolution of current Telco networks towards SDN-NFV (scenario 1) could be “surpassed” by this disruptive approach, creating a “point-of-discontinuity”.

In this case, the full potential of this wave of innovation could be "materialized" into the 5G which could become a sort of border-less “continuum of virtual resources and services” (hosted part in the Network, part in the Cloud but also part at the Edge i.e., up to end-Users terminals, devices, and smart things…). This virtual “continuum” would be capable of enabling new service models (eventually creating new roles for Operators, as well) and new business ecosystems (in industry, agriculture and social contexts). Lean operations, highly flexibility and programmability (via multiple level APIs) will be other main features of such innovative perspective, opening new ecosystems.

Adopting innovation at this level it will be possible enabling the so-called the “servitization” (e.g., anybody and anything ...even a piece of software!... will become a Prosumer of services) , and the “de-perimetrization” of services (e.g., global provisioning of any services where it will be possible for a Service Provider simply renting a physical infrastructure from Infrastructure Providers).