On Mon, Apr 22, 2019 at 12:10 AM Daniel Cegiełka <daniel.cegie...@gmail.com> wrote: > An interesting solution is to keep JFS metadata on a fast separate > ssd. Then on the main disk you have only data structures that can be > recovered more easily.
I find it a terrible idea (for non enterprise deployments) to put the journal on a separate disk. For one, most laptops (which are the "workstations" of today) have only one disk; secondly it would split the data from the journal which would make the movement of the disk containing the data quite problematic (given that a desktop could have more partitions with JFS than actual SSD drives). Ciprian.
