Skip to content

[RFE, GSOC] Improve architecture to scale better for small setups

Currently hyperkitty has all bells and whistles included to scale well for large setups. Using the django ecosystem, there are an async job queue, several database abstractions, and caching opportunities available. For small setups this causes an unfortunate overhead in terms of setup and maintenance. For example, the default ORM polls the database regularly at a high rate and causes a high metadata database load. Moreover, there are additional cronjobs that also handle regular tasks. I propose to merge background task handling to use one dedicated technology and to think about an architecture change in order to reduce overhead for small deployments.

More information is available here:

Stefan

Edited by Stefan Tatschner