Trill: Engineering a Library for Diverse Analytics

Trill is a streaming query processor that fulfills three requirements to serve the diverse big data analytics space: (1) Query Model: Trill is based on the tempo-relational model that enables it to handle streaming and relational queries with early results, across the latency spectrum from real-time to offline; (2) Fabric and Language Integration: Trill is architected as a high-level language library that supports rich data-types and user libraries, and integrates well with existing distribution fabrics and applications; and (3) Performance: Trill’s throughput is high across the latency spectrum. For streaming data, Trill’s throughput is 2-4 orders of magnitude higher than comparable traditional streaming engines. For offline relational queries, Trill’s throughput is comparable to modern columnar database systems. Trill uses a streaming batched-columnar data representation with a new dynamic compilation-based system architecture that addresses all these requirements. Trill’s ability to support diverse analytics has resulted in its adoption across many usage scenarios at Microsoft. In this article, we provide an overview of Trill: how we engineered it as a library that achieves seamless language integration with a rich query language at high performance, while executing in the context of a high-level programming language.