Dr. Dennis Shasha, of the Courant Institute of Mathematical Sciences at NYU, and José Pablo Cambronero, a PhD student at MIT, recently presented the results of their research on AQuery and q/kdb+ at the latest Kx Community NYC Meetup.
AQuery is a simple extension to SQL that makes joins, moving averages, correlations, and other such operations, fast on terabyte-scale data. AQuery unifies the declarative syntax of SQL 92 with the addition of an ASSUMING ORDER clause, moving aggregates, flatten etc to achieve the semantics of kdb+. The system includes an optimizer. AQuery compiles to q.
There are two aspects of this project that Dennis and José presented: (i) AQuery, the language and the optimizer; and (ii) using q/kdb+ as the execution engine for new languages. They provided a series of helpful dos and don’ts.