Re: [PATCH 04/19] sched, numa, mm: Describe the NUMA scheduling problem formally

as per 4) move towards where "most" memory. If we have a large shared memory than private memnory. Why not we just move the process towrds the memory.. instead of the memory moving towards the node. This will i guess be less cumbersome, then moving all the shared memory On Fri, Nov 16, 2012 at 9:5

[PATCH 04/19] sched, numa, mm: Describe the NUMA scheduling problem formally

From: Peter Zijlstra This is probably a first: formal description of a complex high-level computing problem, within the kernel source. Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Cc: Andrew Morton Cc: Peter Zijlstra Cc: "H. Peter Anvin" Cc: Mike Galbraith Rik van Riel Link: http://lk