On 24 Jan 2009, at 17:22, Grant Edwards wrote:
I still can't believe that Windows does it's swapping using a normal filesystem -- and by default it's the same filesystem used for system and application files. It seems like the filesystem code would end up being a serious bottleneck.
> 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. http://lkml.org/lkml/2005/7/7/326 Stroller.