Applications
Characteristics of Event-based systems required by different applications:
- Event identity vs. non-identifiable base events
- Delivery guarantees (from best-effort, at least-once, exactly-once, with/without timing guarantees)
- Granularity of events
- Persistence of events:
- Stateful event aggregation
- Logging of events for auditability and linage
- Lifespan of events
- Event interoperability
- Heterogeneous namespaces
- Combination of QoS properties
- Security and privacy
- Role-based access control
- Pub/Sub on encrypted envelope data
- Scopes
- Support for mobility
- Context-awareness of event systems
- Different degrees (from position, to very rich contexts)
Centralized vs. Distributed Event-Based Systems
- What applications need large broker networks?
Semantics of expressiveness of event systems
- Low-level events, e.g., sensor readings
- High-level events, e.g., corporate actions
- Can we always do bottom-up composition, or can we always break up high level events into lower level ones?
Interfaces to Event-based Systems
- Definition of (domain specific) modeling languages
- Stream visualization in P/S
- Subscription vs. spontaneous querying an event-based system
Guiding exemplary questions:
How can we bring up a world-wide pub/sub network?
Can we run a generic and public pub/sub system on PlanetLab?
What applications would drive the use, the participation, and the adoption of such a system?
Who would/should the users be?
- Pub/Sub researchers, other researchers, application developers, end-users?
What are killer applications for distributed event-based systems?
- Intelligent transportation systems?
- How can event-based systems help build global sensor networks?
- How can small-scale actuator/sensor networks (e.g., in an e-home scenario) profit from pub/sub?
What apps are the low hanging fruit for pub/sub adoption?
- Applications today that are naturally modeled with events, easy to implement, are widely used, would benefit from pub/sub
