Rob Strom

GRYPHON: AN EVENT-BASED PARADIGM FOR MESSAGE BROKERING

 

 

Gryphon is a new distributed computing paradigm for message brokering. Message brokering is the transferring of information in the form of streams of events from information providers to information consumers according to an information flow graph. The information flow graph specifies the selective delivery of events, the transformation of events, and the generation of derived events as a function of states computed from event histories. In this manner, the Gryphon paradigm merges the best features of distributed communications technology and database technology.

Message brokering is an extension of publish-subscribe technology. Our approach, embodied in our experimental system Gryphon, augments the publish-subscribe paradigm with (a) content-based subscription in which events are selected by predicates on their content rather than by pre-assigned subject categories, (b) event transformations, which convert events by projecting and applying functions to data in events, (c) event stream interpretation, which allows sequences of events to be collapsed to a state and/or expanded back to a new sequence of events, and (d) reflection, which allows system managemnet through meta-events. Gryphon technology includes a collection of efficient implementations to support this paradigm and still provide scalability, high throughput, and low latency.

We discuss the Gryphon model, the algorithms we have developed in support of the model, and discuss some of the remaining open research questions.

Site Contents