The Design And Implementation Of A Continuous Query Language

Streams are continuous data feeds generated by such sources as sensors, satellites, and stock feeds. Monitoring applications track data from numerous streams, filtering them for signs of abnormal activity, and processing them for purposes of filtering, aggregation, reduction, and correlation. To efficiently support a variety of real-time monitoring applications, Aurora, Borealis and STREAM are designed and implemented by Brown University, together with Brandeis University and M.I.T., and Stanford University, respectively. Aurora is a general-purpose fully functional data stream management system, which only supports queries in the form of operator network. STREAM formally defines a continuous query language CQL, however, not all functionalities are implemented in STREAM. We are building a data stream management system called Conger, which aims at enabling Aurora to support CQL.Conger is based on aurora. We first design Conger CQL based on Stanford CQL, then we use ANTLR3to parse the Conger CQL that is registered by user. We bind the query parameters with Aurora operators, then generate query execution plan.In order to verify expression ability of Conger CQL, we implement the Linear Road Benchmark (LRB) billing management module. Linear Road Benchmark is developed by the researchers of Aurora and STREAM. Linear Road Benchmark is a data stream management system benchmark. LRB tests show that Conger CQL has rich expression ability that can express complex continuous query.