I read somewhere that Unraid only spins up the specific disk where the data is stored on, thus saving on energy cost but how does Unraid know which disk to spin up?

 

Or is that fact only true when writing data?

Not really looked into it, but I would think that the FUSE file system component that Unraid uses to support User Shares will contain entries that point to where any file accessed that way is really located so I would think the statement is true.   Still an interesting question to ask 😊

Thanks for the quick response!

 

So, is it correct to say that Unraid can spin down all disks to save power, and when accessing data, it will only spin up the specific disk containing the requested data?

Additionally, does a single file always exist on a single disk (excluding the parity disk), ensuring that only one disk is needed for reading a larger file?

Does the same behavior apply when writing data as well?

 

If so, I might switch to Unraid. I've tried achieving something similar with Snapraid and MergeFS, but MergeFS stopped functioning once the disks spun down.

3 hours ago, Skaronator said:

Additionally, does a single file always exist on a single disk (excluding the parity disk), ensuring that only one disk is needed for reading a larger file?

Does the same behavior apply when writing data as well?

With Unraid each disk in the main array is a self-contained file system and Unraid does not support striping so any file must always fit onto a single drive.