There is great excitement about learned index structures, but understandable skepticism about the practicality of a new method uprooting decades of research on B-Trees. In this paper, we work to remove some of that uncertainty by demonstrating how a learned index can be integrated in a distributed, disk-based database system: Google's Bigtable. We detail several design decisions we made to integrate learned indexes in Bigtable. Our results show that integrating learned index significantly improves the end-to-end read latency and throughput for Bigtable.
Cite
@article{arxiv.2012.12501,
title = {Learned Indexes for a Google-scale Disk-based Database},
author = {Hussam Abu-Libdeh and Deniz Altınbüken and Alex Beutel and Ed H. Chi and Lyric Doshi and Tim Kraska and Xiaozhou and Li and Andy Ly and Christopher Olston},
journal= {arXiv preprint arXiv:2012.12501},
year = {2020}
}
Comments
4 pages, Presented at Workshop on ML for Systems at NeurIPS 2020