Learn and build a high-performance vector database engine in C++ integrated with SQLite.
An introduction to building a custom, high-performance vector database engine in C++ integrated with SQLite.
Traditional relational and documentbased databases are designed for exact matching—retrieving records where an identifier matches a specific key or checking if a text field contain...
An analysis of the performance implications of geometric expansion in std::vector, and how pre-allocation minimizes memory churn.
One of the reasons std::vector is the default contiguous container in C is its ability to grow dynamically. Unlike raw arrays, which require compiletime or manual runtime sizing, s...
A beginner-friendly guide to transitioning our vector database to contiguous memory, parallelizing calculations with AVX2 SIMD intrinsics, and analyzing memory-bound performance bottlenecks.
In our previous article, we established a baseline vector search engine using SQLite's Virtual Table API. Our profiling results revealed that 77.64% of the search execution time wa...
