Active complex event processing over event streams

Abstract

State-of-the-art Complex Event Processing technology (CEP), while effective for pattern query execution, is limited in its capability of reacting to opportunities and risks detected by pattern queries. Especially reactions that affect the query results in turn have not been addressed in the literature. We propose to tackle these unsolved problems by embedding active rule support within the CEP engine, henceforth called Active CEP (ACEP). Active rules in ACEP allow us to specify a pattern query's dynamic condition and real-time actions. The technical challenge is to handle interactions between queries and reactions to queries in the high-volume stream execution. We hence introduce a novel stream-oriented transactional model along with a family of stream transaction scheduling algorithms that ensure the correctness of concurrent stream execution. We demonstrate the power of ACEP technology by applying it to the development of a healthcare system being deployed in UMass Medical School hospital. Through extensive performance experiments using real data streams, we show that our unique Low-Water-Mark stream transaction scheduler, customized for streaming environments, successfully achieves near-realtime system responsiveness and gives orders-of-magnitude better throughput than our alternative schedulers.

Probabilistic inference of object identifications for event stream analytics

“…This fits our specific problem well. Second, for the sake of performance, CEP engines are known for their sophisticated capabilities for detecting temporal correlated patterns of events in huge volume event streams [5,12,17,19,36,37]. Thus if we can specify affected events via CEP pattern queries, then the CEP technology can help FISS to detect affected events effectively and efficiently.…”

“…We have implemented all proposed inference techniques in FISS using Java and we take the Active CEP framework [36] as our backend CEP engine. In this section, we present a detailed evaluation of FISS using event streams modeled based on the real-world healthcare system HyReminder [35].…”

“…In recent research on CEP, the focuses have been primarily to enrich the expressiveness of pattern queries and to improve the efficiency of real-time pattern matching [5,17,19,36,37]. We are the first to adopt CEP techniques for probabilistic inference over event streams.…”

Dynamic Aggregate: An Elastic Framework for QoS-Aware Distributed Processing of RFID Data on Enterprise Hierarchy

“…In particular, the processing of data streams has been the primary focus of attention [6], [7]. Over the continuous flow of RFID data stream(s), effective support for complex event processing [8], [9], query processing [6], [10], object tracking and monitoring [11], [12], as well as spatiotemporal data management [13], [14] are crucial to successful RFID implementation. Our research focuses on adaptive aggregation/deaggregation of distributed RFID data based on real-time QoS measure.…”

Transactional stream processing

“…More recently, Wang et al have proposed a stream-oriented transactional model for providing active rule support to complex event processing systems [28]. In this work, active rules monitor continuous query outputs and issue reads and updates to a table, while other continuous queries are concurrently reading from this shared table.…”