Design and evaluation of a wide-area event notification service
ACM Transactions on Computer Systems (TOCS)
Composite Events for Active Databases: Semantics, Contexts and Detection
VLDB '94 Proceedings of the 20th International Conference on Very Large Data Bases
TAG: a Tiny AGgregation service for ad-hoc sensor networks
ACM SIGOPS Operating Systems Review - OSDI '02: Proceedings of the 5th symposium on Operating systems design and implementation
Operator placement for in-network stream query processing
Proceedings of the twenty-fourth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
Self-organizing broker topologies for publish/subscribe systems
Proceedings of the 2007 ACM symposium on Applied computing
IEEE Transactions on Software Engineering
Composite Event Detection in Heterogeneous Sensor Networks
SAINT '08 Proceedings of the 2008 International Symposium on Applications and the Internet
Adaptive and decentralized operator placement for in-network query processing
IPSN'03 Proceedings of the 2nd international conference on Information processing in sensor networks
Composite event detection as a generic middleware extension
IEEE Network: The Magazine of Global Internetworking
| Hi-index | 0.00 |
Communication among devices, including sensors and actuators, forms the basis for ubiquitous computing scenarios such as smart homes. For such scenarios, an event-based tyle of interaction leveraging the publish/subscribe paradigm is well suited as it allows devices to communicate relevant changes of their own state or in environmental conditions as events in a loosely coupled way. Besides individual events, applications often also need information about the occurrence of spatio-temporal patterns of events (called composite events), e.g., to better identify the current situation and, then, be able to take proper actions. While adaptive publish/subscribe middleware has in general been well investigated, only few work deals with making distributed composite event detection adaptive such that the resource usage and the quality of service perceived by applications is optimized even in dynamic scenarios. The reason for this is that adaptive distributed composite event detection requires to tackle an on-line optimization problem that is NP-complete even in the static case. In this paper, we present a self-organizing adaptive strategy for placing event detectors using a heuristic based on a force model. Using an exemplary scenario we show that our approach can significantly reduce the cost associated with event composition, e.g., cost derived from network traffic.