25 July 2014

Software-Defined Infrastructures for Networks, Clouds and Services

NetSoft 2015 will be the first of a series of annual flagship events organised by IEEE Initiative on Software-Defined Networks (SDN) set-up of the IEEE Future Directions Committee to build an world wide community in this area across multiple IEEE societies and councils (http://sdn.ieee.org/about.html). 

What's the main scope of the Conference ? SDN and Virtualization (e.g.  Network Function Virtualization - NFV) could be seen as different expressions (Cloud Computing as well) of an overall "transformation trend", which is deeply impacting and Telecom and IT industries, also in terms of business models and regulation rules. We are already witnessing that this ”softwarization" trend is transforming several other Industries, bringing to optimization processes (and costs) and new values in infrastructures (e.g., 5G). SDN and NFV are creating the conditions to reinvent network and service architectures, to accelerate innovation and to create new business opportunities.

Accordingly, the main theme of NetSoft 2015 is “Software-Defined Infrastructures for Networks, Clouds and Services”. NetSoft is intended to serve as the flagship IEEE forum for technical exchange on software-defined infrastructures and services. NetSoft 2015 aims at providing an excellent forum for the latest innovations, results, and developments in the overall transformation trend, seen also from the business and regulation viewpoints, which is deeply impacting Telecom and IT industries.

The conference will be held on April 13-17, 2015 in London, UK.
The NetSoft 2015 website can be found at: http://sites.ieee.org/netsoft/

Please join us by clicking on the button "Join the  SDN Technical Community" at the link  http://sdn.ieee.org/!

17 July 2014

An OS for SDN meeting Cloud Robotics

They say that innovation, and the development of new industries, is about positioning research on the confluence of distinct streams of emerging technology. Research progress at the intersection of fields is more likely to happen when cross-disciplinary scenarios are encouraged: strategies would include importing ideas from diverse communities and creating an environment enabling to deep cross-collaboration.

I believe that promising opportunities are likely to happen around the intersection of Software Define Networks, Network Function Virtualization - and - Cloud Robotics.

We’re witnessing that the number of smart devices, terminals, machine, things connected to the network is growing day by day. For every cell phone today we have already a few sensors and by the end of this decade there will be more than hundred sensors per each cell phone; for every car we have tenths of microprocessors, the average middle-class household has about 40 microprocessors. The number of connected sensing-computing-actuating objects is increasing tremendously.  This is change of paradigm, as in the future the distinction between the “network” and “what connects to it”, will disappear, becoming these smart devices, terminals, machine, things… like nodes. And this will the case also for Robots.

Let’s go beyond the concept of SDN as the evolution of “networking”, let’s open new perspectives. A Robot, in principle, can be seen as a Software Define – node, capable of providing ICT services and attached to the network. What services ? Imagine the adoption of Robots in agricultural processes, or a wider introduction of smarter Robots in industry (even in hazardous environments improving safety for emergency services, in mining activities, etc) or the use of Robots to support us in our day life activities: a self-driving car is just an example, or the evolution of Jibo or next intelligent personal assistant (like Siri) to evolve to become an intelligent personal Robot.

Devices, machines, smart things or robots will become like new “tools” (enabled by Softwarization) which we’ll have at disposal to develop the Digital Society and the Digital Economy. So, if SDN is seen today as a sort of “network OS”, the 5G OS will be the big technical challenge of the future:  a sort of decentralized OS bringing to radio-fixed convergence beyond the last mile, supporting mobility, embedding security – by design – and spanning from the all Terminals, to the Network to the Cloud.  A key requirement of 5G OS will be allowing QoS with very limited values of “latency” (e.g. in the order of 1-2 ms) for real-time sensing and actuations, but also being developed Open Source to boost innovation!

15 July 2014

SDN-NFV: technology, business sustainability and regulation rules

When looking at future potential deployment scenarios of SDN and NFV, at least three dimensions should be taken into account: technology, business sustainability and regulation rules.

One way to consider the technology variable is the split degree in the (centralized vs distributed) execution of L2-L7 Virtual Network Functions (VNF). In fact, many (but not all) Network Elements (e.g., routers and switches) and Network Functions (e.g., middle-boxes) can be virtualized, i.e., replaced by software running on logical resources (e.g., Virtual Machines). This split will determine different alternatives, as long as the Operations of said VNFs and the services.

In general this could create a split of roles of current Networks and Services Providers (NSP) into “Software- Defined NSP” (SD-NSP) and Infrastructure Provider (IP). A SD-NSP is an Operator owning basically software networks and services platforms, i.e., platforms whose functions (ranging  from L2 to L7) are fully developed, executed and operated in software, in the Cloud and/or  in the Exchanges (centralized vs distributed). SD-NSPs are “borrowing” hardware resources (e.g. just antennas, L0-L1 transmission and processing power) from IPs. An SD-NSP looks like an OTT.

Regarding the business sustainability variable, it is estimated that SD-NSPs will see dramatic costs reductions (e.g. estimation of 40%-50% savings in energy), CAPEX reductions, improved efficiency in the overall Operations (e.g. 25%-35% OPEX savings just by automating processes), reduced time-to-market when deploying services (e.g., quite similar to OTTs today). SD-NSPs will be also potentially capable to “upload and execute” their “software networks and services platforms” anywhere there will be an IP willing to rent hardware and L0-L1 resources.  This is disruptive and it allows to enter new markets with limited investments, e.g., by using just OPEX, or playing as Global Players. Also new service ecosystems (today not sustainable) might be enabled.

Concerning the third variable, regulation rules, the story is even more complex: rules should encourage fair competition, improve the functioning of the market and guarantee basic user rights. In this sense, “mass digitalization and Softwarization” have the potential to create a situation where personal information is created, held and managed at a ‘global’ level, so proper legal frameworks should be put in place.

11 July 2014

Softwarization bringing to a “new order of things”

The Prince” is an impressive political book written by Niccolò Machiavelli, the well-known Italian historian and political theorist. The book was published in 1532, five years after his death: about 5 centuries ago! Machiavelli brilliantly argued in his book that there is no constituency for innovation: "There is nothing more difficult to take in hand, more perilous to conduct, or more uncertain in its success, than to take the lead in the introduction of a new order of things”. A reasoning perfectly applicable to the worries of our world, today!

As a concrete example, this nice piece of article elaborates about the downfall of an Industry, which had to face a technological singularity, a new order of thing, and did not move into “the digital world” well enough and fast enough. Barriers to change culture, hiding innovation, complacency ? A combination of these and other variables, probably…
Taking this metaphor, the technological singularity challenge that we are facing today is that “software is eating the world”: this is due to the ultra-broadband diffusion, the advances in Information Technologies (IT) and optical bandwidth, tumbling hardware costs and availability of open source software. A transformation of economy triggered by technology. In the past, it happened several times…

In fact, software eating the world means “dematerialization”, dramatic changes in processes, and, at the end of the day, a sort of economy phase transition (which, being a non-linear process, it is taking place suddenly, once the critical conditions are reached). You can wait and see, but the problem is that at the transition, it is too late to take any actions: the ecosystem will jump to a new order of things, whatever you’re doing. Nothing strange or surprising: it’s a Nature law, the law governing any ecosystem, even the matter.  That’s symmetry breaking: local decrease of entropy to optimize functionals, but overall global increase. After this sharp transition, a new order of things always emerge, just to quote Machiavelli, till the next one.

It is not probably a question of debating if/when/how software will eat the network (e.g., SDN and NfV are apparently going in that direction). Maybe, the real question is another one: what will be the “new order” after this transition will occur, who will be the winners and how positioning in the new ecosystems, with the right sustainable processes, required for a new flourishing biz ?

09 July 2014

From “Waterfall” towards “Complex Innovation”

“Waterfall” Innovation finished about 15 years ago: the model was moving from research activities to standardization, from systems development by Technology Providers to Service Providers deployments up to services provisioning to Consumers.

Today (and tomorrow) Innovation is likely working the other way round: it is starting from the needs of the Digital Society, from the massive Consumers, Users and from the challenging requirements of the new Digital Economy. Today Could Computing and ICT platforms are easily accessible in any part of the world (almost) on an equal basis. This is making a big difference as it this will reduce the thresholds for several new Players to enter the Digital Economy ecosystems and to provide ICT services for the Digital Society.

In fact, at the end of the day, competition is moving from Capex to Opex, and my guess is that the full transition will happen non-linearly, once a critical threshold will be reached. Just like a phase transition. A system/phenomenon is called linear if effects are proportional to their causes, non-linearity mostly happen when effects are becoming (positive and negative) feedbacks to the input. And that’s what’s happening today: technology development and adoption is accelerating through a systemic combination of several variables. The coming Digital Economy – just like a Complex System - cannot be reduced just to the sum of its parts!

What is the meaning of complexity ? We have to resume the Latin root complexus, which means entwined: a complex system is composed by a network of components variously interconnected (e.g., with positive and negative feedback loops). The study of complex (eco)systems investigates how relationships between components give rise to collective, systemic behaviours of the (eco)system and how the it interacts and forms relationships with its environment. “Waterfall” approaches won't work anymore. 

So, my argue is that Innovation should be deeply embedded in the DNA of Digital Society and Economy, looking at massive markets.That’s why I’m calling it Complex Innovation: a way of making Innovation by dealing with the systemic interdependencies of all socio-economic variables of the hyper-connected world we are living in.

Ultra-broadband diffusion, advances in Information Technologies, increasing optical bandwidth availability, tumbling hardware costs are creating this hyper-connected world and the conditions for the change of paradigm, which is about softwarizing the processes of the Society and the Economy.

06 July 2014

Softwarization paving the way to Cognition-as-a-Service

In Computer Science abstractions hide hardware specific details which are considered not useful at a given level. The OS in fact stands between the Users’ programs and the physical computer hardware, it provides  high level functions (e.g., allocating memory or writing to disk). In this sense, it has been mentioned that SDN can be see seen as a Network Operating System (even if there are interpretation of SDN as a Compiler).

Let’s imagine now to extend this concept (see my previous post) to a distributed “5G Operating System“ capable of spanning from the Terminals, to the Network to the Cloud. Not only that: imagine even embedding cognitive capabilities in the 5G OS, so that cognitive apps can be easily developed by means of APIs, rather than requiring Developers to reinvent the wheel any-time.

Welcome to the world of “Cognition as a Service” (CaaS). By the way, isn’t this also the direction of the next generation cognitive management systems (in the sense of OSS/BSS) that we’re looking for ?

In fact, it will not be possible anymore adopting traditional management approaches (with declared objectives and observed behaviour) for 5G. Complexity and dynamism will be too high. Dynamic or static modeling for (open or closed loop) control will become very complicated and unstable if not supplemented with a variety of methods and control techniques, including (nonlinear) dynamic systems, computational intelligence, intelligent control (adaptive control, learning models, neural networks, fuzzy systems, evolutionary and genetic algorithms), and artificial intelligence.

This trend is already detectable today. It’s not surprising, for example, that a companies like OTTs are pushing the idea of cloud robotics. It is no more Science Fiction when we say that we’ll be living soon with cognitive machines and robots.

03 July 2014

An Operating System from the Terminals to the Network to the Cloud

5G will be the future ubiquitous, ultra-high bandwidth infrastructure: a truly converged “fabric” of IT and Networks resources, where wired and wireless communications will be almost undistinguishable. 5G is expected to become a sort of “nervous system” of the Digital Society and the Digital Economy, capable of bringing an ICT breakthrough to ensure a sustainable economy, to improve quality of life also with new ICT services and to create new jobs.

In order to pursue this, what is needed – in my opinion – is the concrete development and deployment of a distributed Operating Systems (OS) for 5G infrastructures. And it should be based on Open Source. As known, an OS is a “software” managing (for example, in a computer) hardware resources and providing services and abstractions for applications.

Practically Operating Systems acts as an intermediary between applications and the hardware and it is present on almost any device that contains processing and storage, e.g. from cellular phones, to computers, to servers...

Interestingly, there are already activities for developing examples of OS for Software Defined Networks: ONOS in one. ONOS aims at capturing the up-to-date state of the network and maintain a network map; providing an interface for the network control and management applications to update the network map; and programming the forwarding elements, by installing forwarding rules, in response to the changes to the network map by the control and management applications. A similar OS is still missing (at least in part) for the radio infrastructure, where “softwarization” could bring to disruptive scenarios, such as C-RAN, where all functions (e.g., L2-L7) could be moved to Data Centres (e.g., BS pool) interconnected to the RRU via optical backhauling. A then an integration, into a single OS, would complete the breakthrough.

So, the grand technical challenge will be getting this truly distributed “5G Operating System“ capable of spanning from the Terminals to the Network to the Cloud. And a key requirement will be keeping the “latency” in the order of 1 ms, as this will enable a tremendous number of new ICT ecosystems (e.g., imagine providing Cloud APIs for Anything as a Service, including machines, smart things, drones, robots...).