design measures (other than taking backups) to ensure the reliability and integrity of a file system?

Question

I am doing revision for system architecture and came across this question

Suppose that you are designing a particular file system that allows
complex arrangements of file and directories. What design measures
(other than taking backups) might you take to ensure the reliability and
integrity of such a system? 

The only thing I could think of was doing Complete and incremental back ups, which include restore points but the question asks for measures other than back ups. Could someone please let me know other measures that can be taking to ensure reliability and integrity of a filing system?

Answer 1

To add to Rob Conklin's answer, any mirroring (writing to multiple places) is also a method. The mirror can include extra checks to ensure the integrity of the data being written (this is not what RAID usually does, hence the separate answer).

Answer 2

Go look at RAID levels. They are ways of arranging disk storage in ways that allow one or more disks to fail and/or increasing performance by spreading load out over multiple disks.

The fundamental principle is to keep a checksum stored on another disk. This allows a failed disk in an array to be recovered. Different RAID levels allow different numbers of disks to fail without data loss.