A C++ project integrating a Recursive Model Index (RMI) into an LSM-style storage engine. Includes synthetic dataset generation, mmap-backed SST key lists, a learned index inference path, and full benchmarks to compare learned vs traditional index performance.
python performance-engineering benchmarking machine-learning cpp sstable storage-engine data-structures rmi mmap regression-models database-systems systems-programming lsm-tree key-value-store learned-index recursive-model-index cdf-modeling
-
Updated
Nov 25, 2025 - C++