International telecommunication union
Relationship/Coexistence with Mobile edge computing
Download 0.91 Mb.
|
8 Relationship/Coexistence with Mobile edge computing8.1 ICN Mobile Edge ComputingICN is a natural platform to deliver edge services. Its ability to host services, content, with other features such as in-network processing, caching/storage, and multicasting/mobility allows distributed service intelligence and large scale content distribution capability. A distinguishing feature of ICN based edge service platform is contextualized service delivery. Such a platform will manifest the idea of pushing the frontier of computing services and applications away from centralized nodes to the logical extremes of the network. These edge-service realizations can be located in the vicinity of the operator’s Central Office (CO) or at Points-of-Presence (PoP), or eNodeB, or all the way into specific application context such as home networks or transport infrastructure enabling local instantiation of ICN services to consumers, while providing global service delivery through a unified ICN based service control infrastructure. Following points motivates this realization: ICN-MEC– ICN with SDN/NFV integration: Though ICN is an ideal platform for efficient application delivery considering its features to adapt to network disruptions and temporal evolution of services and content, these features also make ICN a complex protocol if it were to be deployed to handle all network functions through distributed control plane mechanisms. Instead ICN should explore design choices offered by NFV/SDN 26 to handle network services like mobility or name resolution functions within practical limits of scalability, which is mostly well handled within local domain scenarios. Further these core ICN functions can be application driven paving way for ICN based service virtualization. Even functions like service chaining27 is more a information-centric operation than host-centric which can be handled by ICN, even in the context of IP flows. Even more service agility via service-chaining can be realized over ICN transport with distributed service functions and dedicated application controllers to aid Interest or Data processing through policy based paths determined by the network services. The co-existance of ICN with SDN/NFV and realizing information-centric service virtualization with rich contextualized content delivery in general is an open research topic. Research Gap: Includes the areas of using base SDN/NFV for ICN slicing while realizing SDN-ICN, NFV-ICN equivalent service planes. Realize ICN Centric Service Virtualization (compute, bandwidth, cache, storage) over NFV-ICN. Realize ICN Centric Network Virtualization 28,29 logical/physical separation of ICN forwarder resources for heterogeneous ICN flows) over SDN-ICN 26 Ravi Ravindran et al, “Towards Software Defined ICN Based Edge Cloud Services”, IEEE, CloudNet, 2013 27 Mayuthan Arumaithurai et al, “Exploiting ICN for Flexible Management of SDN”, ICN, Siggcomm, 2014 28 A. Chanda, C. Westphal, D. Raychaudhuri, Content Based Traffic Engineering in Software Defined Information Centric Networks, in IEEE INFOCOM Workshop NOMEN'13, April, 2013. pdf 29 A. Chanda, C. Westphal, ContentFlow: Mapping Content to Flows in Software Defined Networks, in Proc. IEEE Globecom, December 2013 (arXiv preprint arXiv:1302.1493, January 2013). pdf ICN-MEC – Service Contextualization: : Services are best delivered by locally customizing them to what users want, because users who are located in different locations have different needs and requirements based on their context. Context can be defined as any information that is used to describe the state of an entity and can be classified as being user-, network-, service-, device- centric. Modeling and reasoning of heterogeneous contextual information involves a trade off between complexity of reasoning and expressiveness of data. Mapping the contextual information into service level and network level requirements leads to the challenge of federated and standardized semantic representation of state and context. On the other hand, users themselves access services through heterogeneous devices, network connectivity, with subjective preferences. Moreover, mobility considerations require services to be delivered from the best vantage point. An edge service framework that can handle these dynamic requirements with minimum overhead is desirable. Context of users and services can be inferred and semantics (meaning of data items within a context ontology) of content predicates can be interpreted towards an intelligent dissemination of services and content items. In addition to a physical view of network topology, ICN based services and applications can benefit from a higher level perspective of topology. Research Gap: Generally, ICN research has been limited to study fundamental networking issues related to congestion control, name resolution, multicasting, caching using name-based networking. In-network computing is another ICN dimension which can fundamentally distinguish it from IP networking. This is so because, ICN deals with content objects with metadata associated for security, temporal descriptions, application-centric metadata, which can be correlated with consumer’s intent 30, 32; and conduct immediate transformation if the intent doesn’t satisfy the available content. Modeling and reasoning of heterogeneous contextual information involves a trade off between complexity of reasoning and expressiveness of data31. However the data transformation also has security implications considering the content object based security model. Integration of generic and application-centric in-network functions towards service contextualitzation to aid security function offloading, transcoding, data aggregation and filtering is opegenerally an open area of research. 30 P. Talebifard, R. Ravindran et al, “An Information Centric Networking Approach Towards Contextualized Edge Service “, IEEE, CCNC, 2015 31 P. Talebifard and V. Leung, “A Dynamic Context-Aware Access Network Selection for Handover in Heterogeneous Network Environments”, Proceedings IEEE INFOCOM Workshops, April 10-15, 2011 in Shanghai, P.R. China., pages 385–390. 32 F. Bronzino, S. Stojadinovic, C. Westphal, D. Raychaudhuri, Exploiting Network Awareness to Enhance DASH over Wireless, IEEE CCNC, Las Vegas, January 2016 MEC Service APIs: ICN realizes a new transport using information-centric APIs, this directly elevates the operator as an information-pipe rather than bit pipe provider. Operating an ICN platform, makes the operator an active intermediary to satisfy information requests from consumers. Further service virtualization features can be built over it using control and data plane to handle name resolution, mobility, multicasting etc. From adoption perspective, ICN platform should offer significant technical and business benefits through rich Service APIs to the operatorto services compared to current existing infrastructure considering the the new CAPEX and OPEX in investments towards deploying ICN, which is are still a topic of open Research Gaps: Topics of research include Service-API for ICN deployment leveraging NFV/SDN framework, SLA definitions for Services over an ICN platform 8.2 ICN Based Edge Service Framework: Directory: ITU-T -> focusgroups focusgroups -> International Telecommunication Union focusgroups -> Innovation of ict in Developing Countries focusgroups -> International Telecommunication Union focusgroups -> Telecommunication focusgroups -> Focus Group on Smart Grid focusgroups -> Telecommunication focusgroups -> Telecommunication focusgroups -> Telecommunication focusgroups -> Innovation of ict in Developing Countries focusgroups -> Innovation of ict in Developing Countries Download 0.91 Mb. Share with your friends:
|