From: | Deepak Balasubramanyam <deepak(dot)balu(at)gmail(dot)com> |
---|---|
To: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Learned Index |
Date: | 2017-12-11 08:58:10 |
Message-ID: | CAAerrx8bHiW3rgAqpoLqjhYhk7gHOrDtkE5UC9DAsv1w5FzpEw@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
I came across this paper making a case for indices that use machine
learning to optimise search.
https://arxiv.org/pdf/1712.01208.pdf
The gist seems to be to use a linear regression model or feed a tensor flow
model when a more complicated distribution is needed for the data and allow
SIMD instructions working on top of GPUs / TPUs to speed up lookups. The
speedup observed is anywhere from 40-60%.
That result looks impressive but I don't have enough context on say
rebuilding a neural net on every DML operation. The equivalent operation
that I can relate to on PG would be to rebalance the B-tree for DML
operations.
In your opinion, would a ML model work for a table whose operations are
both write and read heavy? I'd love to hear your thoughts on the paper.
Thanks for reading
- Deepak
From | Date | Subject | |
---|---|---|---|
Next Message | David Rowley | 2017-12-11 09:12:43 | Re: [HACKERS] Removing [Merge]Append nodes which contain a single subpath |
Previous Message | Thomas Munro | 2017-12-11 08:56:41 | Re: Parallel Index Scan vs BTP_DELETED and BTP_HALF_DEAD |