Continuous query processing applications; Stream-oriented applications
Streaming applications typically involve the processing of continuous data streams for the purposes of filtering, aggregation, correlation, transformation, pattern matching and discovery, and domain-specific temporal analytics. These applications often require such continuous processing to be performed with both high throughput and low latency and are able to tolerate approximate results and forego some of the persistence requirements of standard database transaction processing applications.
A large fraction of streaming applications is monitoring oriented: they involve the tracking of events or activities to identify and act upon situations (or patterns) of interest, either manually or automatically. This so-called “sense and respond” model requires query results to be generated in real time (meaning low latency) as results lose their utility over time. As such, persistence of all the...
- 1.Options Price Reporting Authority (OPRA) Traffic Projections, http://www.opradata.com/specs/projections_2005_2006.pdf