On 2009-01-24, Stroller <[email protected]> wrote: > > 3. Does creating the swapfile on a journaled filesystem (e.g. > > ext3 or reiser) incur a significant performance hit? > > None at all. The kernel generates a map of swap offset -> disk > blocks at swapon time and from then on uses that map to perform > swap I/O directly against the underlying disk queue, bypassing all > caching, metadata and filesystem code.
I supposed that the NT kernel does something similar. One implication of that is that the filesystem is then not allowed to move blocks around if they are part of an "active" swap file? Not that I'm aware of filesystems that shuffle blocks around while they're part of an open file, but one might imagine something like that happening as part of some sort of balancing algorithm. -- Grant
