Topic Description
From the beginning of the information era, data processing and communication were two distinct facets of the whole ICT world, each one evolving on its own, and developing specific and often closed/proprietary technologies. However, the last two decades have seen huge progresses in computing power at affordable prices and exponential growth of connectivity needs for any kind of users (business/residential, fixed/mobile, etc.). The advent of highly-distributed information processing paradigms, such as Cloud Computing and Internet of Things, has urged the ICT industry to break the boundaries between computing and communication, to the point that vendors that were traditionally active in building network equipment are now offering also advanced computing hardware platforms, and vice-versa.
The current industry trend of convergence between processing and networking ecosystems clearly shows that software will play an unprecedented dominant role also in future communication environments [1]. Computing, storage, and connectivity services, as well as any other present and future application instances, will be deployed in the form of virtualized assets within a software-defined infrastructure running on top of general-purpose processing and communication hardware, all managed and made available under the cloud “as a Service” paradigm.
Such a progressive “softwarization” of the communication infrastructure started around 20 years ago with the first attempts to develop flexible and reconfigurable (i.e., programmable) transmission and networking functions, e.g., Software Defined Radio [2] and Active Networks [3], respectively. However, only recently the software “revolution” in networking has reached significant levels of interest from both academia and industry, owing to the development of key technologies such as Software-Defined Networking (SDN) [4], emerging solutions such as Network Function Virtualization (NFV) [5], and perspective paradigm shifts such as the ones enabled by Network Operating Systems envisioned in 5G scenarios [6].
This technological convergence and infrastructure sharing between the computing and communication systems portend a scenario with a “fog” of micro-clouds composed of generalized virtual functions providing both applications and network services that supplement those deployed in traditional cloud datacenters. Such virtual functions are being deployed in distributed clusters of small-scale datacenters typically located in current Points of Presence at the network edge. However, it is envisioned that such small-scale datacenters will evolve into a dispersed number of nano-datacenters, scattered throughout the network with indistinguishable communication, storage and computing systems sharing the same hardware devices, i.e, commodity servers [7]. Moreover, thanks to virtualization properties and self-contained service abstraction, an end-to-end service can be addressed as a composition of virtual functions, including also network services required to process the traffic flows.
Consequently, it is of prime importance to devise orchestration mechanisms that facilitate the live deployment and lifecycle management of these soft elements, at the application level, the server level, and the network level within a single domain and across multiple domains. Without such orchestration it will not be possible to enable dynamic establishment of generalized virtual functions chains, according to service requirements. Orchestration capabilities can also include context management features to enforce context-aware adaptation strategies and negotiate QoS-aware delivery policies with the underlying layers [8].
Such scenario poses several challenges and requirements, especially in terms of flexible and efficient mechanisms for automated provision and management of generalized virtual functions on top of software-defined infrastructures. These challenges are many-fold, with many open questions that need to be addressed in the areas of:
- network “softwarization,” which requires unified management of computing, storage, and network resources for the effective deployment, lifecycle management, and run-time configuration of generalized virtual functions;
- abstraction models and open standard interfaces, needed for assuring vendor interoperability;
- adaptation and optimization mechanisms, which must be enforced at global and/or local level for coping with user demand, application requirements, resource unavailability, etc.
- K. Pretz, “Software Already Defines Our Lives - But the impact of SDN will go beyond networking alone,” IEEE The Institute, Vol. 38, No. 4, p. 8, December 2014.
- F.K. Jondral, “Software-defined radio: basics and evolution to cognitive radio,” EURASIP Journal on Wireless Communications and Networking, Vol. 2005, No. 3, pp. 275-283, August 2005.
- A. Galis, et al., “A flexible IP active networks architecture,” Active Networks, Lecture Notes in Computer Science, Vol. 1942, pp 1-15, Springer Berlin Heidelberg, 2000.
- N. McKeown, et al., “Openflow: enabling innovation in campus networks,” ACM SIGCOMM Computer Communication Review, Vol. 38, No. 2, pp. 69–74, April 2008.
- M. Chiosi, et al., “Network Functions Virtualisation,” White paper at the SDN and OpenFlow World Congress, ETSI, Tech. Rep., 2012.
- S. Shin, et al., “Rosemary: A robust, secure, and high-performance network operating system,” ACM SIGSAC CCS 2014, Scottsdale, AZ, USA, pp. 78-89, November 2014.
- A. Manzalini, R. Minerva, F. Callegati, W. Cerroni, A. Campi, “Clouds of Virtual Machines in Edge Networks,” IEEE Communications Magazine, Vol. 51, No. 7, pp. 63-70 , July 2013.
- F. Paganelli, M. Ulema, B. Martini, “Context-aware service composition and delivery in NGSONs over SDN,” IEEE Communications Magazine, Vol. 52, No. 8, pp. 97-105, August 2014.
