Cost Model

From: neto brpr <netobrpr(at)gmail(dot)com>
To: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Cost Model
Date: 2017-12-20 17:29:37
Message-ID: CALhhvc3MVe59VavoeE3ahrMeO6dh6fh9fkEzOT-tf-Vfp6qm-g@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hello All

Anyone know if there is a patch that changes the PostgreSQL cost model in
some way?
I'm working with an tuning technique, based in hybrid storage environments
(HDDs and SSDs).

I need to know if exist any patches that allow postgresql to differentiate,
for example, I/O costs of read and write and also costs of sequential read
and random read.

This is because the read and write speeds of SSDs are asynchronous. On SSDs
the read speed can be 340 times faster than writing.
In Hard Disks the sequential reading and writing values are almost the
same. Also in HDDs the random reading is much slower than the sequential
reading.
In the SSDs the sequential and random reading are practically the same, due
to being a purely electronic device.
I would like to know if there is any solution related to this that I said,
because I would not need to develop this if something similar already
exists.

It would be interesting if there is a patch that inserts new parameters in
the postgresql.conf file to:

- sequencial read page cost
- random read page cost
- sequencial write page cost
- random write page cost

Any comment, hint about it or something, please inform me.

Regards,
Neto.

<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Livre
de vírus. www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>.
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2017-12-20 17:34:31 Re: Schema-qualify function calls in information_schema
Previous Message David Fetter 2017-12-20 17:21:22 Schema-qualify function calls in information_schema