George Pavlou

Probabilistic in-network caching for information-centric networks

In-network caching necessitates the transformation of centralised operations of traditional, overlay caching techniques to a decentralised and uncoordinated environment. Given that caching capacity in routers is relatively small in comparison to the amount of forwarded content, a key aspect is balanced distribution of content among the available caches. In this paper, we are concerned with decentralised, real-time distribution of content in router caches. Our goal is to reduce caching redundancy and in turn, make more efficient utilisation of available cache resources along a delivery path. Our in-network caching scheme, called ProbCache, approximates the caching capability of a path and caches contents probabilistically in order to: i) leave caching space for other flows sharing (part of) the same path, and ii) fairly multiplex contents of different flows among caches of a shared path. We compare our algorithm against universal caching and against schemes proposed in the past for Web-Caching architectures, such as Leave Copy Down (LCD). Our results show reduction of up to 20% in server hits, and up to 10% in the number of hops required to hit cached contents, but, most importantly, reduction of cache-evictions by an order of magnitude in comparison to universal caching.

Modelling and evaluation of CCN-caching trees

Networking Named Content (NNC) was recently proposed as a new networking paradigm to realise Content Centric Networks (CCNs). The new paradigm changes much about the current Internet, from security and content naming and resolution, to caching at routers, and new flow models. In this paper, we study the caching part of the proposed networking paradigm in isolation from the rest of the suggested features. In CCNs, every router caches packets of content and reuses those that are still in the cache, when subsequently requested. It is this caching feature of CCNs that we model and evaluate in this paper. Our modelling proceeds both analytically and by simulation. Initially, we develop a mathematical model for a single router, based on continuous time Markov-chains, which assesses the proportion of time a given piece of content is cached. This model is extended to multiple routers with some simple approximations. The mathematical model is complemented by simulations which look at the caching dynamics, at the packet-level, in isolation from the rest of the flow.

Cache less for more in information-centric networks

Ubiquitous in-network caching is one of the key aspects of information-centric networking (ICN) which has recently received widespread research interest. In one of the key relevant proposals known as Networking Named Content (NNC), the premise is that leveraging in-network caching to store content in every node it traverses along the delivery path can enhance content delivery. We question such indiscriminate universal caching strategy and investigate whether caching less can actually achieve more . Specifically, we investigate if caching only in a subset of node(s) along the content delivery path can achieve better performance in terms of cache and server hit rates. In this paper, we first study the behavior of NNCs ubiquitous caching and observe that even naive random caching at one intermediate node within the delivery path can achieve similar and, under certain conditions, even better caching gain. We propose a centrality-based caching algorithm by exploiting the concept of (ego network) betweenness centrality to improve the caching gain and eliminate the uncertainty in the performance of the simplistic random caching strategy. Our results suggest that our solution can consistently achieve better gain across both synthetic and real network topologies that have different structural properties.

A management and control architecture for providing IP differentiated services in MPLS-based networks

As the Internet evolves toward the global multiservice network of the future, a key consideration is support for services with guaranteed quality of service. The proposed differentiated services framework is seen as the key technology to achieve this. DiffServ currently concentrates on control/data plane mechanisms to support QoS, but also recognizes the need for management plane aspects through the bandwidth broker. In this article we propose a model and architectural framework for supporting DiffServ-based end-to-end QoS in the Internet, assuming underlying MPLS-based explicit routed paths. The proposed integrated management and control architecture will allow providers to offer both quantitative and qualitative services while optimizing the use of underlying network resources.

An overview of routing optimization for internet traffic engineering

Traffic engineering is an important mechanism for Internet network providers seeking to optimize network performance and traffic delivery. Routing optimization plays a key role in traffic engineering, finding efficient routes so as to achieve the desired network performance. In this survey we review Internet traffic engineering from the perspective of routing optimization. A taxonomy of routing algorithms in the literature is provided, dating from the advent of the TE concept in the late 1990s. We classify the algorithms into multiple dimensions: unicast/multicast, intra-/inter- domain, IP-/MPLS-based and offline/online TE schemes. In addition, we investigate some important traffic engineering issues, including robustness, TE interactions, and interoperability with overlay selfish routing. In addition to a review of existing solutions, we also point out some challenges in TE operation and important issues that are worthy of investigation in future research activities.

IP ROUTING ISSUES IN SATELLITE CONSTELLATION NETWORKS

SUMMARY The growth in use of Internet-based applications in recent years has led to telecommunication networks transporting an increasingly large amount of Internet Protocol (IP)-based traffic. Proposed broadband satellite constellation networks, currently under development, will be required to transport IP traffic. A case can be made for implementing IP routing directly within the constellation network, in order to transport IP traffic well and to provide good support for IP multicast and for emerging IP-based Quality of Service (QoS) guarantees. This paper examines strategies for implementing and operating IP routing effectively within satellite constellation networks, given known constraints on the constellation resulting from satellite mobility, global visibility, routing and addressing.

Cache less for more in information-centric networks (extended version)

Ubiquitous in-network caching is one of the key aspects of information-centric networking (ICN) which has received widespread research interest in recent years. In one of the key relevant proposals known as Content-Centric Networking (CCN), the premise is that leveraging in-network caching to store content in every node along the delivery path can enhance content delivery. We question such an indiscriminate universal caching strategy and investigate whether caching less can actually achieve more. More specifically, we study the problem of en route caching and investigate if caching in only a subset of nodes along the delivery path can achieve better performance in terms of cache and server hit rates. We first study the behavior of CCNs ubiquitous caching and observe that even naive random caching at a single intermediate node along the delivery path can achieve similar and, under certain conditions, even better caching gain. Motivated by this, we propose a centrality-based caching algorithm by exploiting the concept of (ego network) betweenness centrality to improve the caching gain and eliminate the uncertainty in the performance of the simplistic random caching strategy. Our results suggest that our solution can consistently achieve better gain across both synthetic and real network topologies that have different structural properties. We further find that the effectiveness of our solution is correlated to the precise structure of the network topology whereby the scheme is effective in topologies that exhibit power law betweenness distribution (as in Internet AS and WWW networks).

Admission control for providing QoS in DiffServ IP networks: the TEQUILA approach

Admission control is a key component for QoS delivery in IP networks because it determines the extent to which network resources are utilized and whether the contracted QoS characteristics are actually delivered. Unlike most of the existing approaches, we adopt a holistic view and position the admission control in a unified architecture, where traffic engineering and service management interact in different levels of abstraction and timescale. Distinguishing between service subscription and invocation epochs, the applied admission logic is based on a feedback model, the operation of which can be tuned by operational policies and business strategies. The validity of the approach is supported by a scalability analysis while it has also been verified by a prototype system.

On management technologies and the potential of Web services

From the early days of management research and standardization in the late 1980s, there has always been a quest for a management technology that would be general enough to be used for network, system, service, and distributed application management; efficient in terms of information retrieval time, computational resources required, and management traffic incurred; and easy to use in order to reduce development time and operational costs. From protocol-based approaches such as OSI management and SNMP, attention shifted to distributed object and Web-based approaches. Recently, XML-based approaches and, in particular, Web services have been emerging as a promising Internet-based technology that might also be used for management. In this article we survey first the key aspects of protocol and distributed object approaches to management. We subsequently examine Web services as a distributed object approach to management, and consider its suitability, usability, and performance in comparison to SNMP and CORBA.

Curling: Content-ubiquitous resolution and delivery infrastructure for next-generation services

CURLING, a Content-Ubiquitous Resolution and Delivery Infrastructure for Next Generation Services, aims to enable a future content-centric Internet that will overcome the current intrinsic constraints by efficiently diffusing media content of massive scale. It entails a holistic approach, supporting content manipulation capabilities that encompass the entire content life cycle, from content publication to content resolution and, finally, to content delivery. CURLING provides to both content providers and customers high flexibility in expressing their location preferences when publishing and requesting content, respectively, thanks to the proposed scoping and filtering functions. Content manipulation operations can be driven by a variety of factors, including business relationships between ISPs, local ISP policies, and specific content provider and customer preferences. Content resolution is also natively coupled with optimized content routing techniques that enable efficient unicast and multicast- based content delivery across the global Internet.