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

From Andrey Borodin
Subject Re: ML-based indexing ("The Case for Learned Index Structures", a paper from Google)
Date
Msg-id DEDD8D37-B39B-4624-BF47-4F8C62A2C22E@yandex-team.ru
Whole thread Raw
In response to Re: ML-based indexing ("The Case for Learned Index Structures", a paper from Google)  (Stefan Keller <sfkeller@gmail.com>)
Responses Re: ML-based indexing ("The Case for Learned Index Structures", a paper from Google)  (Peter Geoghegan <pg@bowt.ie>)
List pgsql-hackers

> 20 апр. 2021 г., в 22:56, Stefan Keller <sfkeller@gmail.com> написал(а):
>
> Are there any advances in a learned index for PostgreSQL?

BTW take a look into PGM [0]. I'm slowly working on implementing it.
I think it is kind of straightforward to implement it as extension.
I've started from forking B-tree[1]. I've removed support of anything that is not int4.
Then I plan to remove opclass\comparator abstraction layer.
And then add interpolation search over the page before binsearch.
And, maybe, add delta compression. Or, perhaps, fix WAL, which is completely broken. Or add some vectorisation.
The concurrency is from a regular B-tree, of cause.

This is my pet project. So, unsurprisingly, I've done only parts of big plan :) I would appreciate any help.

Thanks!

Best regards, Andrey Borodin.


[0] https://pgm.di.unipi.it/
[1] https://github.com/x4m/pg2m


pgsql-hackers by date:

Previous
From: Ondřej Žižka
Date:
Subject: Re: Synchronous commit behavior during network outage
Next
From: Tom Lane
Date:
Subject: Re: when the startup process doesn't