On 10/13/07, Carsten Otte <[EMAIL PROTECTED]> wrote: > No method is in place to ever change a backing. Thus, a page will > never be moved from one swap slot to a different swap slot. This is > why linux is not very good at keeping swap usage local, and is not > good at using low prio swap for hot pages.
I don't think that were implied by what was suggested. It would even be a bad idea to move pages through swap (because you might cause them to page in just for re-ordering them). What would make a difference is the way slots are allocated on swap devices. As I mentioned between the lines of my posts, some of the apparent design criteria do not apply to VDISK (or even mainframe disk I/O in general). But that algorithm is probably in the generic part of Linux. The right way would probably be to add a way to steer the algorithm. In that case an option on swapon to select between first-fit and best-fit, for example. And while working on my pony, I would also like an API on the block device tell the driver to drop a block from backing store (also COW devices would enjoy that). Rob -- Rob van der Heij Velocity Software, Inc http://velocitysoftware.com/ ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390