Tidy Tuples and Flying Start: fast compilation and fast execution of relational queries in Umbra

Abstract Although compiling queries to efficient machine code has become a common approach for query execution, number of newly created database system projects still refrain from using compilation. It is sometimes claimed that the intricacies generation make compilation-based engines too complex. Also, major barrier adoption, especially interactive ad hoc queries, long compilation time. In this paper, we examine all stages execution and show how reduce overhead. We incorporate lessons learned decade generating in HyPer into design manages complexity yields high speed. First, introduce framework establishes abstractions manage complexity, yet generates single fast pass. Second, present program representation whose data structures are tuned support Third, new compiler backend optimized minimal compile time, simultaneously, superior performance competing approaches, e.g., Volcano-style or bytecode interpretation. implemented these optimizations our Umbra it possible unite execution. Indeed, achieves unprecedentedly low latencies. On small sets, even faster than interpreter like DuckDB PostgreSQL. At same on large its throughput par with state-of-the-art HyPer.