Welcome to Nanyan Jiang's Homepage

Overview

Overview | Associative Rendezvous (AR) |Cascading Local Behavior (CLB)| Contributions

Overview:

Pervasive environments are information-rich environments, but the information has more locality and commonality than the Internet in general.   We propose Associative Rendezvous (AR) as a paradigm for content-based decoupled interactions for pervasive applications. By doing so, we abstract away the time sequences of device readings and actuator actions into a simple interface that resembles flows with reactions.  Underneath this simple abstraction, however, are some powerful protocols that convey information throughout the wireless network system, together with internet backbones, information repositories and computing centers.  We use Meteor, a content-based middleware infrastructure to support AR interactions.

Supporting these applications requires an application programming and management paradigm where the behaviors as well as the interactions of applications elements (components, devices, services, etc.) are dynamic and context, content and capability aware

The model builds on content-based discovery and routing services and defines associative rendezvous as an abstraction for content-based decoupled interactions.

Associative Rendezvous (AR):  

Associative Rendezvous (AR) is a paradigm for content-based decoupled interactions with programmable reactive behaviors. The AR interaction model consists of three elements: Messages, Associative Selection, and Reactive Behaviors.

AR message: An AR message is defined as the triplet: (header, action, data). Symmetric post primitive

Associative Selection: profiles stored locally are evaluated (matched) by incoming message.

Reactive behavior: The action field of the message defines the reactive behavior at the rendezvous point. Basic reactive behaviors currently defined include store, retrieve, notify, and delete.

Characteristics of AR interaction:

  • Content-based decoupled interactions:

    • All interactions are based on content, rather than names or addresses (associative)

    • The participants (e.g. senders and receivers) communicate through an intermediary, the Rendezvous point

    • The communication is asynchronous. The participants can be decoupled both in space and time.

  • Programmable reactive behaviors:

    • The reactive behaviors at the Rendezvous points are encapsulated within messages

      • flexibility, expressiveness, and multiple interaction semantics

Cascading Local Behavior:

In the Cascading Local Behaviors programming model, the behaviors of individual application elements (i.e., components, services) are locally defined in terms of local state, and context and content events, and result in data and interest messages being produced. Interactions, compositions and application flows emerge as a consequence of the cascading effect of such local behaviors, without having to be explicitly programmed.

Contributions:

Associative Rendezvous interaction extends the conventional name/identifier based rendezvous in two ways. First, it uses flexible combinations of keywords (i.e. keywords, partial keywords, wildcards, ranges) from a semantic information space, instead of opaque identifiers that have to be globally synchronized. Second, it enables the reactive behaviors at rendezvous points to be embedded in the message or message request. AR differs from emerging publish/subscribe paradigms in that individual interests (subscriptions) are not used for routing and do not have to be synchronized - they can be locally modified at a rendezvous node at anytime.


Meteor is designed and developped at Rutgers University's TASSL lab at the CAIP Center