With the Internet becoming the ubiquitous vehicle for all forms of communication, in this project we argue that the time has come to enhance the original address-to-address communication primitive. With this regard, Information Centric Networking (ICN) is a new paradigm in which the network layer provides users with access to content by names, instead of providing communication channels between hosts. ICN infrastructure addresses content by names without reference to their location, route a user request, which includes a destination content-name, toward the closest copy (original server, cache, end-users) of the content with such a name and deliver the content back to the requesting host(s). This is believed to offer several advantages, including: improved efficiency, thanks to in-network caching and content-based routing; simplified handling of mobile and multicast communication; ability to work in a non-infrastructure mode; peer-to-peer communications support; a content-oriented security model; content-oriented access control/QoS; network awareness of transferred content. However, there are several issues of ICN that still need to be addressed: i) Identification of killer applications; ii) the complexity and scalability of the proposed naming and routing functionality; iii) the cumbersome support for push services; iv) several security and privacy concerns; v) the need to devise a credible migration path from the communication paradigm of the current network infrastructures; vi) routing efficiency; vii) applicability in scenarios such as the aftermaths of a disaster.
GreenICN has therefore planned to take an application driven solution design to drive the design on the GreenICN architecture, the application and middleware layer APIs and the specific network layer solutions. We believe that multiple application requirement driven approach will help in the design of generic solutions that can not only handle the considered applications, but potentially future applications too. The two exemplary application scenarios that are under consideration are: i) Disaster scenarios: The aftermath of a disaster e.g., hurricane or tsunami, when energy and communication resources are at a premium and the ability to exploit fragmented networks with only intermittent connectivity and to efficiently utilize energy is very important; ii) Video Scenarios: A key requirement for video scenarios is the need for a scalable and efficient dissemination mechanism, an efficient mobility (especially source/group) and caching (especially CDN like services) solutions. Recently, we have identified a new application — Name based Service Chaining in SDN — that has the potential to be a killer application.
The requirements derived from these applications were used to design specific solutions for efficient routing, caching, mobility, pub/sub and etc. Moreover, this helped in the design of the APIs between the application layer and the network layer as well as the APIs to the energy layer. The APIs are designed to form an interface between the network functionality and the required application layer functionality. For instance, topic based publish/subscribe can be performed at the application layer with enhanced support from the network layer. The application layer can take care of matching keywords based subscription/publication to relevant topics and the network layer can support with efficient data transfer and caching.
Summary of Work Done
The main focus of GreenICN is to improve NDN/CCN in key areas such as energy efficiency, function in disaster scenarios (fragmented/disconnected operations, pub/sub: interest for something published in the future: useful to receive early warnings) and video scenarios (pub/sub for video, mobility and migration). We have also identified a new use-scenario and also proposed “Server assisted” architectural enhancements to NDN/CCN. In order to receive feedback as well as to contribute the design of ICN/NDN, we have actively pursued IETF, IRTF, ITU and MPEG and have achieved publications in peer reviewed top conferences (41), workshops (21), Poster/Demonstration (16), Journals (20) and Books/Chapters (1). Additionally, we have about 90 technical-reports/invited-talks available to the public and furthermore GreenICN members made 84 non paper related presentations at conferences (keynotes), panels, universities, standardization and ICT events.
Routing, Forwarding and Traffic control
We worked on topics pertaining to i) End-to-End traffic/resource control; ii) Network-side traffic/resource control; and iii) Traffic/resource control for Prioritization. The work on End-to-End traffic/resource control has two complimentary proposals. The first proposal develops a control protocol (SAID) to enable applications to effectively use a name-based one-to many information dissemination architecture and can be used for both publish/subscribe as well as query/response congestion control. The second proposal titled In-Network Resource Pooling Principle looks at in-network router caching from a different angle and questions whether caches can take on alternative roles and assist with the control of network congestion and in-network resource management. The work on Network-side traffic/resource control has two proposals, MTTE and retransmission control for fragmented ICN. MTTE dealt with energy saving by shutting down unused links, whereas the retransmission control mainly focused on reducing waste of bandwidth and energy consumption due to retransmission in fragmented networks. The work on Traffic/resource control for prioritization has two proposals, NREP and PMTTE. NREP mainly focused on the aftermaths of a disaster scenario where the network could be fragmented and that resources could be scarce. PMTTE, on the other hand, focused on differentiation of priority traffic with tree-based traffic engineering based on MTTE. On the resource control front, we proposed an i) efficiency enhancing FIB lookup scheme for CCN-based network realms; ii) an enhancement to ICN to improve flexibility for service chaining in an SDN environment; iii) we present a node architecture for routing/forwarding enhancements to the CCNx core for efficient group communication; and iv) provide an overview of our mobility extensions to routing and forwarding mechanisms.
End Systems and Middleware
GreenICN has developed a specification of
- High Level API which expose a set of high level methods for Applications to call;
- Middleware with low computing and memory demands, high energy efficiency, potentially portable to all platforms and exposing the High Level API
- Low Level API which are called by the Middleware e.g., to call local computing resources, get energy information, request security functionalities and access Face;
GreenICN has contributed to a book chapter highlighting the advantages of ICN for energy efficient communication that argues that energy optimization techniques applied on the current Internet infrastructure will not result in orders of magnitude increase in energy efficiency due to the limitations posed by the design and the need to consider the possibility of deploying future Internet architectures, namely Information Centric Networking. We then attempt to answer the question “Does caching really reduce the energy consumption of the entire network?”. Additionally, as part of this work, we presented energy consumption models for an individual ICN router and an overall network that consists of ICN routers. These models are used to evaluate whether the GreenICN architecture satisfies the energy requirements. Moreover, we also developed a novel tree-based traffic energy efficient control mechanism for ICN, where the number of links used is minimized without causing congestion. Furthermore, an efficient FIB lookup scheme for improving routing and energy efficiency in a CCNx-based network-realm has been designed and evaluated. Our work on prioritization in a disrupted networks also has the potential to save energy and we plan to investigate this further.
Aftermaths of Disaster
We created two network scenarios to clarify the WP2 success criteria, i.e., i) 95% connectivity in fragmented networks applied to physical damaged area; ii) and 40% power reduction applied to power loss area. We further studied two technologies, collaborative communications and cell zooming, for reduction of power consumption for disasters. We also developed an energy consumption model for a LTE-based BS. Further, we designed a protocol which seamlessly integrates a Publish/Subscribe mechanism, i.e., COPSS, and collaborative communication in order to deliver crucial information to multiple recipients in shelters of a black out area in an energy efficient manner. This protocol is designed so as to interwork with a LTE-based cellular network and to enable efficient group communication mechanisms for efficient rescue operations. We also added a new routing table in the CCN architecture in order to be able to maintain and resolve information when the network gets fragmented. We further investigated the performance of mule-based communication under network fragmentation and given an ICN environment. We have also designed an algorithm to estimate the popularity of interests in a fragmented network and aggregate requests according to that estimation/indication. Routing and relaying was also considered under the newly proposed DID framework, which is also taking into account priorities for certain data items. On the security front, we investigated security mechanisms depending on network status, meaning solutions that adapt procedures and performances to the status of the network; for instance, after a disaster some pre-defined encrypted information could be released; the challenge is in designing distributed solutions that do not necessarily depend on interactions with central servers/authorities that may be unavailable / unreachable. Further, we proposed Identity-Based Aggregate Signatures (IBAS), an extension to IBS, to reduce the signature overhead. Furthermore, our approach for decentralized authentication has been further developed, generalized, and thoroughly evaluated. Finally, a comparison of energy efficiency of the proposed vs. existing authentication procedures was further investigated and the difficulties have been much better put into focus.
We have adopted a bottom-up approach for the design of the video-sharing framework and have identified the main components of the framework. We have built on three basic concepts of sharing, namely synchronization of streams in caches for efficient video delivery on mobile devices, collaborative video streaming over 3G and P2P, and 3G offloading through Wi-Fi access points to provide initial context for the technical considerations of this design. Based on this we have identified the main building blocks of the framework, namely video coding, descriptor-based Pub/Sub, flow control, multipath scheduling, cache management and optimization, mobile-to-mobile communication management, routing and mobility management. To consolidate our video-sharing framework, relations among components are studied, e.g., video coding is considered in mobile-to-mobile communication protocol and the feature of video content is exploited in optimized caching. Further, power consumption of mobile networks, terminals, and applications are measured, and a theoretical framework for the assessment of the Quality of Experience of users streaming video in urban railway networks was proposed. We have improved COPSS to increase its efficiency as a publish/subscribe delivery system. We have also worked on the design and evaluation of a “Location-Independent Routing Layer” (LIRA), which is a proposal for a smooth migration from the IP Internet to an ICN-based future network. Furthermore, we proposed On-Path Resolver Architecture (OPRA), a mechanism to support producer mobility as well as scalable routing.
Expected Final Results and Impact
GreenICN is designed by putting applications at its forefront. This has two folds advantage: 1) the applications present their own challenges and allows the GreenICN architecture to evolve to handle such use cases; 2) the chosen applications are based on real problems currently faced in Japan, Europe and many parts of the world and therefore would generate a huge interest in the results and the product of this research. Therefore we believe that there will be significant impact from the design, implementation and deployment results of GreenICN in the research community as well industry. The impact in Europe, Japan and across the world will be high. Dependability will be enhanced by the scalable and secure, name-based communications infrastructure that provides increased flexibility in mobility, resource control and network manageability. Further, distributed in-network storage enables robust content delivery. Thanks to the increased efficiency and reliability of the network, disaster recovery, health-care, and surveillance applications can be deployed in large scale, as they can substantially benefit from such a communications infrastructure. The reduced cost and complexity of the application interface will be significant reasons for more widespread deployment. A major impact of such deployment will be improvements in the way new services and service infrastructures will be developed and operated. Today, it is still a challenge to develop a new service, let alone a new service infrastructure, mostly due to the low levels of abstraction provided by current networks (middleware systems try to raise this level of abstraction somewhat, but are only partially successful due to their inherent efficiency penalties). GreenICN research is concerned with ICN will allow application developers to better abstract away location and address related issues for information delivery, so that developers can concentrate on developing service logic. These developments will create new business opportunities and increase the quality of life and economic activity for European and Japanese citizens.