Analysis and Optimization of Cache Resource Allocation for Energy-Efficient Information-Centric Networking

Lead Research Organisation: Liverpool Hope University
Department Name: Faculty of Sciences


It is predicted that Internet video streaming and downloads will account for more than 76 percent of all consumer Internet traffic in 2018. The tremendous growth of multimedia traffic has given rise to the demand for highly scalable and efficient content retrieval and dissemination in the Internet. However, the Internet was originally designed to enable host-to-host communication and lacks natural support for content distribution. In this context, Information-Centric Networking (ICN) has emerged as a new paradigm for future Internet, where the network interprets, processes, and delivers name-identified content to the users independently of the host location. ICN deploys in-network caching that enables content to be retrieved from multiple locations to achieve low dissemination latency and network traffic reduction.

Serving as its fundamental building block, efficient in-network caching is vitally important for ICN. The distinct features of in-network caching such as transparency, ubiquity and fine-granularity have made traditional caching theory, models and optimization approaches inapplicable to ICN caches. Therefore, significant research efforts have been devoted to tackling the very challenging problem of in-network caching. The existing research works have been primarily focused on the simulation studies of ICN caching. However, analytical modelling of ICN cache networks is indispensable for the understanding of the intrinsic behaviors and features of in-network caching. The analytical models reported in the current literature for ICN mainly adopt unrealistic assumptions, such as independent reference model and unknown chunk-level object popularity, and are commonly based on the inefficient Leave Copy Everywhere (LCE) cache decision policy only. Furthermore, due to both increasing energy cost and CO2 emission, energy efficiency of networks and systems becomes a dramatically growing concern. Consequently, energy-efficiency of ICN has also been investigated by some studies, which are mainly based on unrealistic models of topology and content requests. To the best of our knowledge, analytical modelling and optimization of cache resource allocation for energy-efficient information-centric networking with transparent, ubiquitous and fine-granular caches has not been reported in the existing literature.

This project will investigate in-network cache resource allocation to achieve energy-efficient and timely content dissemination in the context of Information-Centric Networks. To tackle this challenging problem progressively, our work will be focused on three major tasks: 1) design of an intelligent cache decision policy with low complexity for ICN to reduce cache redundancy, increase the cache diversity and leverage the correlation between content requests; 2) development of novel analytical tools for evaluating the energy efficiency and performance of the proposed cache decision policy in terms of cache hit ratio and request response time with multimedia applications and heterogeneous network conditions; 3) development of a centralized optimization algorithm to investigate the impact of traffic conditions and network environments on the efficiency of cache allocation and a distributed cache allocation scheme that allocates appropriate cache locations of content chunks to minimize the energy consumption. The insights into energy-efficient cache allocation obtained in the aforementioned Tasks 1 and 2 will be feed into the distributed management scheme design in Task 3. The research proposed in the project is believed to among the first of its kind on the analysis and optimization of in-network cache allocation for energy-efficient ICN. The implications of this research will contribute directly to ICN in-network caching in both theoretical and practical sides and pave the way for future green Internet with multimedia applications.

Planned Impact

Outcomes of this project have the potential to deliver scientific and technological advances that will strengthen the UK's competitiveness in future Internet, energy-efficiency, and many other related areas, thus contribute to the economy and wellbeing of the UK. Internet Service Providers (ISPs), content providers, networking equipment manufacturers, and Internet services and applications need to collaborate to advance ICN and its applications for planning and development of future products and services. Potential beneficiaries of this project include:

Internet Users
The Internet has been transformed from a host-to-host communications network to a global infrastructure for the massive distribution of information. According to recent predictions, global IP traffic has been and continues to be exponentially skyrocketing. Internet users are increasingly interested in receiving information wherever it is located and whenever it is needed, rather than in connecting a particular host or server. By decoupling information from its sources, the novel ICN approach is expected to realize highly scalable and efficient distribution and replication of large amounts of data to provide excellent Quality-of-Experience (QoE) for Internet users.
The proposed research on energy-efficient in-network caching is a vital and necessary component of ICN. The research outcomes can help bring ICN enabled effective and efficient content dissemination closer to practice. Successful deployment of content dissemination applications will offer Internet users best QoE in a cost-effective way, by minimizing the request response time and optimizing network resources. Various content dissemination applications can be tested and implemented in both near and medium time scales, such as medical devices and healthcare management systems, home media sharing, scalable and efficient video delivery, etc. Internet users will also benefit from the ICN on energy-efficient content distribution. Energy-efficiency is a growing concern due to both increasing energy costs and CO2 emissions. As more and more users access the Internet using mobile devices equipped with batteries, it will be desirable to adopt energy-efficient content distribution to prolong the battery life.

Internet industry
The exponential growth of data demand poses a significant challenge because the Internet was designed as a communications network rather than a data distribution network, which impacts every part of the Internet ecosystem and the whole industry needs novel and economical ways to solve these problems. A possible solution to provide scalable and efficient content distribution is to develop information-centric future Internet. In moving to energy-efficient ICN, ISPs, content providers, equipment manufacturers, and Internet services and applications will find themselves at an advantage in terms of networks and energy cost, QoE, and breath of services, which will undoubtedly enhance their competitiveness and profitability.

Academic Beneficiaries
Academic beneficiaries include the researchers carrying out related research. They will benefit from the theoretical and methodological advances to be achieved. Research outcomes will be disseminated for the network research community and other disciplines, to obtain a better understanding and keep track of cutting-edge research in this emerging area. The proposed project endeavors to enhance the knowledge and skill base of the UK through training and developing researchers.

Outcomes of this project may contribute to Internet industry and standardization activities, aid government agencies and other bodies to obtain a better understanding of the potential impact of ICN, thereby appropriate regulatory policies can be investigated. The contributions to the research community, internet industry, and standardization bodies will lead to a subsequent impact on the increase of quality of life and economic activity for the general public.

Related Projects

Project Reference Relationship Related To Start End Award Value
EP/M013936/1 01/04/2015 08/09/2015 £92,017
EP/M013936/2 Transfer EP/M013936/1 01/12/2015 30/06/2017 £88,210
Description A new analytical model was developed to gain valuable insight into the caching performance of Information-Centric Networking (ICN) with arbitrary topology and bursty content requests. The analytical model can be used as an efficient tool to investigate the impact of key network and content parameters on the performance of caching in ICN. Moreover, we propose a cost-aware caching scheme to study the Quality-of-Service (QoS) and cost of ICN and investigate the inner association between them. A multi-objective evolution algorithm is adopted to find the optimal cache resource allocation. We also propose a novel proactive content caching algorithm based on machine learning without centralised training data to improve the cache hits rate while preserving users' privacy. (one paper is ready to be submitted)

A category based context aware and incentive based reputation mechanism was proposed to defend against the internal attacks to enhance the big data veracity in mobile cloud computing (MCC).
Exploitation Route Built upon our research findings, others can further develop more advanced ICN caching models, optimization methods and security mechanisms and apply them to tackle important challenges in 5G architecture and Cyber-Physical systems (CPS). The analysis and optimization tools for ICN caching developed in the proposed research could enable the developments of various applications such as scalable video delivery, home media sharing, and etc.
Sectors Digital/Communication/Information Technologies (including Software),Other