Thrash reduction no longer a priority for Linux kernel devs?

Standard

Version 3.5 of the Linux kernel has been released.

freshly installed ipod linux, booting. during ...

freshly installed ipod linux, booting. during Wikipedia:Workshop Köln (Photo credit: Wikipedia)

One of the changes it includes is the removal of the “swap token” code – one of the very few ‘local’ memory management policies grafted on to the ‘global’ page replacement mechanisms in the kernel.

There are various technical reasons offered for the removal of the code – on which I am not qualified to comment – but the borrow line is that it was broken in any case, so the removal seems to make sense.

What does slightly disturb me, though, is the comment that Rik Van Riel, the key figure in kernel memory management code, makes:

The days of sub-1G memory systems with heavy use of swap are over.
If we ever need thrashing reducing code in the future, we will have to
implement something that does scale.

I think the days of sub-1G systems are far from over. In fact I suspect there are more of them, and more of them running Linux, than ever before and that trend is not going to stop.

He’s right of course about the need to find that code that works – my own efforts (in my MSc report) didn’t crack this problem, but I do think there is more that can be done.