How to improve load times for issues with a large number of articles?

Describe the issue or problem
I have a journal (actually conference proceedings) which has about 280 articles per issue distributed over a large number of sections (i.e. conference sessions). Naturally the loading time for a single issue index page is quite long (about 22 sec).

Does anyone have a comparable issue or any idea how to improve the loading times for the issue index page?

What application are you using?
OJS 3.3.0-4

Hi @ronste1,

There have been some improvements made starting with 3.3.0-15 that should help; see for details:

Once you’re up to date on the latest stable release, if you’re still having problems, let’s dig into it further here.

Regards,
Alec Smecher
Public Knowledge Project Team

Thanks @asmecher ,

I’ll try. Actually I had a typo above. The instalaltion is at 3.3.0-14 already. So a small step for installation and hopefully a large step for performance. :wink:
I’ll let you know how it goes.

1 Like

Hi @asmecher ,

to give you some feedback. Going to 3.3.0-16 resulted in a “minor” improvement of about 2 sec. Well, its 10% so probably not that minor. But eventually I was able to identify the main origin of the long loading times. The customized template included calls to fetch statistical data (number of downloads and views). The time to fetch this data obviously scales with the number of articles.
Fast enough for smaller issues but 280 articles obviously were too many.

1 Like

Hi @ronste1,

Thanks for following up! We’ve got increased capabilities for jobs and offline processing in 3.4.0, so it should be possible to regenerate stats on e.g. a daily basis and leave the templates to just present the data. You could also cache that data in your own code; have a look e.g. at the oldGregg theme plugin.

Regards,
Alec Smecher
Public Knowledge Project Team

This topic was automatically closed after 27 days. New replies are no longer allowed.