Re: Bloom Indexes - bit array length and the total number of bits (or hash functions ?? ) !

From: Fabien COELHO <coelho(at)cri(dot)ensmp(dot)fr>
To: Avinash Kumar <avinash(dot)vallarapu(at)gmail(dot)com>
Cc: PostgreSQL Developers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Bloom Indexes - bit array length and the total number of bits (or hash functions ?? ) !
Date: 2019-06-08 06:11:03
Message-ID: alpine.DEB.2.21.1906080759310.9244@lancre
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


Hello Avinash,

> I was testing bloom indexes today. I understand bloom indexes uses bloom
> filters. [...]
>
> So the question here is -
> I assume - number of bits = k. Where k is the total number of hash
> functions used on top of the data that needs to validated. Is that correct
> ? If yes, why do we see the Index 1 performing better than Index 2 ?
> Because, the data has to go through more hash functions (4 vs 2) in Index 1
> than Index 2. So, with Index 1 it should take more time.
> Also, both the indexes have ZERO false positives.
> Please let me know if there is anything simple that i am missing here.

You may have a look at the blog entry about these parameters I redacted a
few year ago:

http://blog.coelho.net/database/2016/12/11/postgresql-bloom-index.html

--
Fabien.

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Justin Pryzby 2019-06-08 06:38:58 Re: Should we warn against using too many partitions?
Previous Message Arthur Zakirov 2019-06-08 04:16:49 Re: [PROPOSAL] Drop orphan temp tables in single-mode