Welcome to Geeklog Tuesday, October 17 2017 @ 02:54 pm EDT


Status: offline

dinsdale

Forum User
Newbie
Registered: 04/01/2005
Posts: 7
It's not obvious from the geeklog documentation the performance hits geeklog can get from the accumulation of articles stored in the database. The geeklog archive scheme makes a big difference when utilized. It effectively divides articles into two piles: the current topics pile and the archived pile. The smaller the first pile, the greater the performance boost. To make it smaller, you shift articles to the archives pile -- a topic dedicated to archiving (see geeklog documentation).

The procedure normally is to set an expiration date during the editing phase of an article. You have the option to automatically archive the article or delete it at the selected expiration date. However, if you neglect to follow this step during the editing phase and years pass and you have 30,000 articles in current topics, there's another way to archive the older material. What I decided to do was a bit of data manipulation using the sql client.

The variable of interest in the stories table was statuscode. The three values this variable takes are 0 (current), 10 (archive), and 11 (vaporize). I decided to use the date variable to target rows by year. The update statement might look like this:

update gl_stories set statuscode = 10 where date < "2010%";

This finds all articles from 2009 to the dawn of civilization and changes the value of statuscode to 10 in those cases the value is not this number.

This procedure can make a big difference in geeklog performance depending on the load you tend to carry in terms of article content.

Of course, backup your db before you attempt this.

Status: offline

1000ideen

Forum User
Full Member
Registered: 04/08/2003
Posts: 1289
I wonder if the developers have an alternative idea as I don`t archive my stories.

Status: offline

Dirk

Site Admin
Admin
Registered: 12/01/2002
Posts: 13073
Location:Stuttgart, Germany
Quote by: dinsdale

It's not obvious from the geeklog documentation the performance hits geeklog can get from the accumulation of articles stored in the database.


I'd be interested to see some numbers to back this up. How many stories are we talking about?

bye, Dirk

Status: offline

dinsdale

Forum User
Newbie
Registered: 04/01/2005
Posts: 7
As I mentioned in the item above, there were over 30,000 stories of varying lengths, usually a few paragraphs. Mind you, when you click on the archived articles, you get the same slowness, but they're old, news anyway.

All times are EDT. The time is now 02:54 pm.

  • Normal Topic
  • Sticky Topic
  • Locked Topic
  • New Post
  • Sticky Topic W/ New Post
  • Locked Topic W/ New Post
  •  View Anonymous Posts
  •  Able to post
  •  Filtered HTML Allowed
  •  Censored Content