Re: elevator algorithm considered irrelevant

Mark Lord (mlord@pobox.com)
Wed, 18 Nov 1998 13:53:29 +0000


Rik van Riel wrote:
> I know what case: the case where people have installed Linux
> on one partition with the swap partition a bit out of the way.

Mmm... As my fuzzy memory recalls, this was the exact situation
that was helped enormously by the minimal head movement logic.

Swap accesses tend to come in pairs: write something out,
and then read something in to replace it. Long sequences of such.

Leaving the heads over the swap area for the duration is a huge
win under such circumstances, as opposed to doing continuous
time-consuming long seeks back and forth from swap to data to swap
to data to..

But there is a theoretical chance for starvation.

-- 
mlord@pobox.com

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/