Our PGM-index has been featured on Hacker News! We warmly thank everyone who showed interest and shared our achievements over the Web!
We are also happy to announce that our library now supports (i) indexing files on disks, and (ii) orthogonal range queries in k-dimensions.
https://t.co/0ymYyOBYSR very excited about this. If the claims are true, this is going to be be very disruptive and “game changing”, even.
— Mark Papadakis (@markpapadakis) January 25, 2021
The PGM-index - New smaller faster index structure for databases https://t.co/ZKEpidSPsG
— Peter Zaitsev (@PeterZaitsev) January 26, 2021
Fascinating! Check it out, another learned index data structure; this time without neural nets, The PGM-index. Presentation: https://t.co/Xxw9ZlBc5U Paper: https://t.co/4VXcVrizmV
— The Real Databass 🐟 (@therealdatabass) January 25, 2021
Piecewise Geometric Model index (PGM-index) is a data structure enables fast lookup, predecessor, range searches & updates in billions using orders of magnitude less space than traditional indexes while providing the same worst-case query time guarantees.https://t.co/cWqYR3RJjA
— Kelly Sommers (@kellabyte) January 25, 2021
Piecewise Geometric Model index (PGM) for databases: Pls. validate and benchmark this promising new database index and on a real #DBMS like #PostgreSQL 🙂! /cc Paolo Ferragina and Giorgio Vinciguerra @Unipisa https://t.co/M74egAc86q
— Stefan Keller (@sfkeller) January 25, 2021
The Piecewise Geometric Model index (PGM-index) looks extremely nice https://t.co/J5rtn06a3q On the page you can find links to the paper and a presentation.
— Miguel Ángel Pastor (@miguelinlas3) January 25, 2021
New Find: PyGM is a Python library that enables fast query operations on sorted number lists w/ a tiny memory overhead. Internally, it uses PGM-index, a state-of-the-art learned data structure that scales to billions of elements in a few tens of MBs.https://t.co/7zhlkxZQLj
— Ian Maurer (@imaurer) January 25, 2021