We have some information regarding performance scattered across documentation. However, for customers with larger installations we should have a dedicated guide on performance collating all relevant points in a single location.
Repo for notes and ideas:
- Collect Wikis, doc pages, tickets, etc. giving recommendations re. performance and scalability.
- Can we plug into srhodes's auditing platform for checking performance best practices on a project basis?
- Notes from initial discussion.
- From this GSheet:
- We have seen customers invariably need to tune the memory and cache size of their Magnolia instance and lacking the expertise to do so. The documentation states recommended values that are not upheld in the example code. Example: Best Practices Documentation states "Using a magnoliaAuthor.setenv.memory.maxPercentage value of 60 (the default value)", but the example code given in https://docs.magnolia-cms.com/paas/deployment/webapp_deployment.html has 80% configured. This leads to customers needing to tune a value that they have never touched. https://docs.magnolia-cms.com/paas/best-practices/memory_resources.html
- We don't have any scalability metrics on Magnolia PaaS.
We have a scaling guide for on-prem but we should have sizings for infrastructure to show customers best practise. Some information is available from the best practices chapters of the PaaS documentation.
- According to awarinner DAM problems are mostly about sizing and growing out of an internal JCR based DAM sooner than expected. We need to take this into account when writing the large scale content guide (i.e. by emphasising the importance to not underestimate future growth).
- See the comment from Jan 9-th 2024 from mdrapela for some more links to existing documentation to review.
- New monitoring endpoints for monitoring repository and query statistics:
- Collect and document useful performance related metrics
- List of tickets of our initiative to include in the documentation (those that have Documentation update required checked).