C++ Search for Database Kernels
Built In, Not Bolted On
If you need a search engine in a C++ product, your options are surprisingly thin. There's no well maintained, feature-rich C++ search library. Some projects have even shipped an entire JVM just to use Java Lucene. Meanwhile, an open-source Apache 2.0 C++ search engine supporting full-text, vector, geo, and semi-structured search has been quietly hiding inside the database kernel since 2015 and almost nobody knows it exists.
IResearch was built to be embedded directly into database kernels. First powering ArangoSearch, now serving as the search foundation of SereneDB, it's a fully featured search engine functionally on par with Lucene and faster. This talk is about how it was built, how it performs, and what ten years of C++ development taught us.
The takeaway: the C++ ecosystem has its own feature rich and production ready search engine. Attendees will leave with practical insights on writing SIMD-friendly C++ for sparse data, an honest look at maintaining a decade-old C++ codebase and a library they can actually use in their products.
Andrey Abramov
Andrey is the Founder and CTO of SereneDB, where he is building a real-time analytical search and OLAP database. With over 15 years of experience in C++, Andrey specializes in production-grade search engines and database kernel internals.
Prior to SereneDB, Andrey was the mastermind behind ArangoSearch, a search engine natively integrated into ArangoDB's distributed multi-model database core.
Earlier in his career, Andrey held senior engineering roles at EMC and Quest Software, where he managed teams and led the development of enterprise-scale systems.