On Wed, 24 Jan 2001, Chuck Cranor wrote:

	> >possibily of readahead) methods the designer(s) found?
	> 
	> the advantage is that you can always do big cluster single i/o
	> operation pageouts (provided the swap disk isn't too fragmented).   
	> that makes recovery from heavy anon memory usage pretty fast (much
	> faster than the old VM).
	> 
	> the disadvantage is that to do readahead pagein you will have
	> to do multiple I/O operations since the data will not be contig
	> on the disk.

	That means big cluster pageout has the price of not always being able to
	do contiguous swapin readahead.

	I'm considering doing the same for Linux 2.5 swap scheme, though I'm not
	sure how that will affect the system in practice. 

Since paging behavior is by inherently unpredictable, readahead
is of quite limited use for paging.  To do readahead you need to
evict pages you know were used in favor of pages you don't know
will ever be needed.

If you're swapping entire processes in and out of memory, 
readahead would be useful.  But in that case you should be able
to cluster the pages for one process during swapout because 
you should not be going through the standard pager.

Eduardo