Request a Demo
Home   >   mailing lists

Re: ML-based indexing ("The Case for Learned Index Structures", apaper from Google) - Mailing list pgsql-hackers

From Oleg Bartunov
Subject Re: ML-based indexing ("The Case for Learned Index Structures", apaper from Google)
Date
Msg-id CAF4Au4x1VFpytRKkkeRmaxbiyk-xomNZm+V3+xcksNrCDY+p-w@mail.gmail.com
Whole thread Raw
In response to ML-based indexing ("The Case for Learned Index Structures", a paperfrom Google)  (Nikolay Samokhvalov <[email protected]>)
Responses Re: ML-based indexing ("The Case for Learned Index Structures", apaper from Google)
List pgsql-hackers
Tree view 
On Mon, Dec 11, 2017 at 11:11 PM, Nikolay Samokhvalov
<[email protected]> wrote:
> Very interesting read: https://arxiv.org/abs/1712.01208
>
> HN discussion: https://news.ycombinator.com/item?id=15894896
>
> Some of the comments (from Twitter
> https://twitter.com/schrockn/status/940037656494317568): "Jeff Dean and co
> at GOOG just released a paper showing how machine-learned indexes can
> replace B-Trees, Hash Indexes, and Bloom Filters. Execute 3x faster than
> B-Trees, 10-100x less space. Executes on GPU, which are getting faster
> unlike CPU. Amazing."
>
> Can those ideas be applied to Postgres in its current state? Or it's not
> really down-to-earth?

Oleg made some analysis of the paper.

pgsql-hackers by date:

Previous
From: Alexander Korotkov
Date:
Subject: Re: CUBE seems a bit confused about ORDER BY
Next
From: Robert Haas
Date:
Subject: Re: [HACKERS] Proposal: Local indexes for partitioned table