Section: Research Program

Algebraic Foundations for Query Optimization and Code Synthesis

We consider intermediate languages based on algebraic foundations for the representation, characterization, transformations and compilation of queries. We investigate extensions of the relational algebra for optimizing expressive queries, and in particular recursive queries. We explore monads and in particular monad comprehensions and monoid calculus for the generation of efficient and scalable code on big data frameworks. When transforming and optimizing algebraic terms, we rely on cost-based searches of equivalent terms. We thus develop cost models whose purpose is to estimate the time, space and network costs of query evaluation. One difficulty is to estimate these costs in architectures where data and computations are distributed, and where the modeling of data transfers is essential.