Skip site navigation (1) Skip section navigation (2)

Re: Database theory question

From: "Mag Gam" <magawake(at)gmail(dot)com>
To: "Jan Danielsson" <jan(dot)m(dot)danielsson(at)gmail(dot)com>
Cc: pgsql-novice(at)postgresql(dot)org
Subject: Re: Database theory question
Date: 2008-04-20 18:33:33
Message-ID: 1cbd6f830804201133l27a6ba8cy18517928e22c8026@mail.gmail.com (view raw or flat)
Thread:
Lists: pgsql-novice
Thanks for the responses.

I am planning to run complex queries such as when a event changes
drastically. I also plan on generating several types of reports pertaining
the data. I was wondering if there was a generic but powerful schema for
time series related stuff.





On Sun, Apr 20, 2008 at 5:59 AM, Jan Danielsson <jan(dot)m(dot)danielsson(at)gmail(dot)com>
wrote:

> Mag Gam wrote:
> > While reading this article, History tables and event logging --
> > http://www.ibm.com/developerworks/web/library/wa-dbdsgn2.html, I
> realized I
> > try to do event logging in SQL.
> >
> > My question are: Is SQL a good tool for event logging? Does anyone have
> a
> > sample table sctucture for the most optimal way of event logging?
> Current I
> > have 2 tables. 1 table with timestamps, another with event. Can anyone
> > recommend a better way?
>
>    It depends on what you want to accomplish in the end. Just storage,
> or do you want to perform "complex" queries?
>
>   I store my firewall & sshd authtentication attempt logs in a
> postgreql database. And I've written a python based web front-end to
> view the data, and pull out statistics and such. I also have a "Generate
> abuse report" link for when there's been a break-in attempt.
>
>   Personally, I see no reason to separate timestamps and event entries,
> unless you're getting a lot of events during the same timestamp, and
> want to save some space. But there are other factors to consider: How
> many events do you get per timestamp? How long times does an insert into
> a single table vs two tables take?
>
>   You should probably try the simplest possibly solution first, and see
> if you need to make it more complex as you can gather empirical data on
> how it performs.
>
> --
> Kind regards,
> Jan Danielsson
>
>
>

In response to

pgsql-novice by date

Next:From: Nils ZierathDate: 2008-04-21 10:21:20
Subject: Re: Remote / network connection fails
Previous:From: Jan DanielssonDate: 2008-04-20 09:59:59
Subject: Re: Database theory question

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group