Help with big data sets

[maxphilippov]

I've set up Sveltia with a self-hosted Gitea. I have 30 000 records. Git doesn't struggle with this amount of files, git status and git commit are fast. But the initial load of Sveltia CMS takes up to 6 minutes. It caches afterwards, but that's just not acceptable for the team I'm working with. I need to reduce it to something like 20 seconds.

From what I see: Sveltia gets all the files with content in batches, I tried upping the batch size but that just makes it slower to load each batch, the overall time spent on loading doesn't change. I'd expect loading just the filenames (so I can individually request contents later) would not take more than 20 seconds. From what little time I've spent looking at the code I assume full contents of a file is required for sorting and searching.

So the question is: is there a setting I'm missing to turn off searching and sorting so I can just write a custom backend file and load only file names and then just load every file on demand when I open the markdown editor? Or it's all pretty tied up together? I'm fine with not having search at first (and later rolling my own solution) and maybe sort based on last change when requesting it from Gitea.

[kyoshino]

See https://sveltiacms.app/en/docs/intro#scalability

Sveltia CMS (and Git-based CMSs in general) is best suited for small to medium-sized projects. For large projects that involve thousands of entries and assets, consider using a more robust CMS solution designed for enterprise use cases.