Are the scheduled tasks needed?

[Dandroid5000]

What is the reason for the Library Analysis and Clustering being ran on a schedule?

I ask because I've just completed the initial Analysis and Clustering inside the AudioMuse UI and it took a very long time - it took days on my mini-PC and large music collection. I worry that with these being ran on a schedule, that my machine would be left in a constant state of running CPU-intensive tasks. Is that a reasonable concern?

After the initial Analysis, do subsequent runs re-scan my entire music library, or does it only re-scan new files and those that have been modified?

Thanks.

p.s. I'm really impressed with the results of the Instant Mix feature now. Great work on this!

[NeptuneHub]

Hi,
Thanks for reaching us.

You can off course just try it but:

• analysis task analyze only the new song. If you re run a new analysis, it skip the song already analyzed.
• Clustering re-start from zero, so it takes “the same time every time”. And it useful to re-create the cluster including new song. It doesn’t takes day and you can decide to do less clustering RUN to takes less time. Or to not run clustering and just enjoy the other functionality that run synchronous.

On Kubernetes you can set a maximum resources usage in your deployment if you don’t want to use all the cpu.

I usually run the analysis each night, so in this way it includes just the few new song and in this way is very fast. The clustering still on night but maybe weekly.

[Dandroid5000]

Ok cool. If the analysis skips songs that have already been analyzed then there's no issue keeping the jobs running on a schedule.
Thank for confirming

[NeptuneHub]

Yes analysis is designed to skip the already analyzed track. Even because if for any reason you interrupt it, you can restart without re-doing the whole analysis.