From: "Warner, Gary, Jr" <gar@uab.edu>
To: postgres performance list <pgsql-performance@postgresql.org>
Subject: Big Memory Boxes and pgtune
Thread-Topic: Big Memory Boxes and pgtune
Thread-Index: AQHSMTIiezg8uYvCuEaMEHgAkS4lPw==
Date: Fri, 28 Oct 2016 15:44:13 +0000
Message-ID: <0cfccbc24b9e480a9452bf9ca79e96e4@UABEX4.ad.uab.edu>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

I've recently been blessed to move one of my databases onto a huge IBM P8 c=
omputer.   Its a power PC architecture with 20 8-way cores (so postgres SHO=
ULD believe there are 160 cores available) and 1 TB of RAM.

I've always done my postgres tuning with a copy of "pgtune" which says in t=
he output:

# WARNING
# this tool not being optimal
# for very high memory systems

So . . . what would I want to do differently based on the fact that I have =
a "very high memory system"?

Suggestions welcome!

(There are several different databases, mostly related to our work in socia=
l media and malware analytics.   The databases are smaller than available R=
AM.   Around 80 million social media profiles with 700M or so links, growin=
g by 10% a week or so.  The malware database has extracted statistics and d=
ata about 17 million malware samples, growing by about 5% per week.   The S=
ocial Media side has just shy of 100 'fetchers' that insert/update (but don=
't delete.)   A few human analysts interact with the data, hitting some pre=
tty heavy queries as they do link analysis and natural language processing,=
 but mostly the database use is the "fetchers")

---------------------------------------------------------------------------=
---
Gary Warner, Director CETIFS      gar@uab.edu
Center for Emerging Technology Investigations Forensics & Security
The University of Alabama at Birmingham (UAB)
205.422.2113
---------------------------------------------------------------------------=
---


--=20
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance