Developing Information Networking Further: From PSIRP to PURSUIT

Abstract

Abstract. PSIRP (Publish-Subscribe Internet Routing Paradigm) is an EU FP7 funded project that has developed a clean-slate architecture for the future Internet, based on the publish-subscribe primitives (rather than the send-receive ones), all the way down to the core networking functions. The PSIRP vision is a pure information-centric Internet architecture, possibly providing remedies to many of the current Internet problems. In PSIRP, all is information and everything is about information. Content-based identities, recursive application of ideas, cryptographic techniques, and the Trust-to-Trust principle are all extensively used to achieve the design goals. Furthermore, incentive compatibility and socio-economic considerations are guiding the design from the outset, to ground the project in reality and to provide credible and viable potential deployment paths. The project has developed, implemented, and preliminarily evaluated solutions for rendezvous, topology formation and routing, and information forwarding, with ongoing work currently focusing in experimenting. A new (also EU FP7 funded) follow-on project, PURSUIT (PublishSubscribe Internet Technologies), will refine and further explore and expand PSIRP's vision. We believe that this will eventually lead to a more complete architecture and protocol suite, thereby providing for more extensive performance evaluation and investigations on scalability. This paper provides an overview of the PSIRP concepts and the developed architecture, along with some key results, and outlines the research directions of the PURSUIT project, focusing on the project goals and its expected outcomes.

Proactive selective neighbor caching for enhancing mobility support in information-centric networks

“…Hence, they can be appropriately implemented using the receiver-driven (pull-based) communication primitive in the Content-Centric Networking (CCN) architecture [8], exploiting CCN's ability to effectively disseminate that same information to multiple interested receivers. Similarly, it can be implemented in the Publish-Subscribe Internetworking (PSI) [6] architecture, which supports a receiver-driven model at the rendezvous (or resolution) layer. Moreover, one-to-many dissemination can exploit both CCN and PSI's native multicast capabilities.…”

MAP-Me: Managing Anchor-Less Producer Mobility in Content-Centric Networks

“…Similarly, NetInf [3] and JUNO [33] report network mobility events to a resolution service, which may incur in network load in case of high mobility [34]. PURSUIT [10] instead uses a rendezvous system to handle network mobility, which requires notification to the topology manager at each handoff and, in some cases, the re-computation of the forwarding identifier used to compute the path to the producer, affecting the handoff delay [40], [34]. Finally MobilityFirst [31] uses a global name resolution service (GNRS), which is updated when a node changes point of attachment.…”

Popularity and gain based caching scheme for information-centric networks

“…In order to solve these problems, researchers have focused attention on research of new Internet architecture [2]. Information-centric networks (ICN) emerges at the right moment and has received extensive research and development, among which; the respective schemes under this background include the contentcentric networking (CCN)/ named-data networking (NDN) [3], data-oriented network architecture (DONA) [4], NetInf [5], PDX [6], etc. In ICN, each node is equipped with a built-in cache, and its core idea is that storing content through internal routers of the network, then the user requests forward in the network.…”