Towards a Secure Rendezvous Network for Future Publish/Subscribe Architectures

Abstract

Abstract. Publish/Subscribe is often regarded as a promising paradigm for Future Internet architectures. Its information oriented nature and its particular security features have stimulated current research efforts which aim at applying publish/subscribe principles to a clean-slate Internet architecture. One of the core components of publish/subscribe architectures is the rendezvous network. Any security failure that a rendezvous network may face will probably jeopardize the operation of the whole (inter-)network. In this paper we highlight security requirements and potential security issues of rendezvous networks and we present security solutions that can be applied in order to shield them.

A cache-aware routing scheme for information-centric networks

“…For example, in the PURSUIT architecture [7] the mediation system consists of the topology management function [31], responsible for constructing the delivery tree between publisher(s) and subscriber(s), and the rendezvous function [32], responsible for matching publisher(s), subscriber(s), and content objects. The mediation system can either determine the shortest path or a path based on predefined traffic engineering objectives [13].…”

Access control enforcement delegation for information-centric networking architectures

“…Access control issues have been surmounted so far using cryptographic solutions in information naming or at the packet level [4]. Nevertheless these solutions simply transfer the problem of access control to the endpoints or to the rendezvous point [3], whereas our system leverages the role of in-network mechanisms.…”

Context-aware information delivery in assistive environments over a publish-subscribe internet

“…A method to achieve this is described in [12]. As a by-product this avoids or limits (D)DoS attacks (except to the rendezvous network [13] that needs special attention).…”

“…The publishers do not usually hold references to the subscribers, neither do they know how many of these subscribers are participating in the interaction and similarly, subscribers do not usually hold references to the publishers, neither do they know how many of these publishers are participating in the interaction (space decoupling). The fact that the publication-subscription matching takes place at an independent point in the network along with the time/spacedecoupling, reduces synchronization requirements between the participating entities and thus allows for efficient support of multicast, mobility, as well as multihoming and indirection [13].…”