Re: clustering page-ins

Chuck Lever (cel@monkey.org)
Thu, 12 Aug 1999 21:32:37 -0400 (EDT)

On Thu, 12 Aug 1999, Jamie Lokier wrote:
> Which gives me a thought. Is there's an algorithm to discover the
> optimal random-access readaround cluster size, analogous to the
> sequential readahead window algorithm?

it might also help trond's nfsv3 work if there was a per-file cluster size
that was initialized to a system default (64k, or 8k, or get_maxreadahead)
and then varied as necessary.

i think you'd need to maintain some kind of data structure for each
cluster in a file, like a bitmap or histogram, to watch how many pages
were faulted in each cluster.

- Chuck Lever

--
corporate:	<chuckl@netscape.com>
personal:	<chucklever@netscape.net> or <cel@monkey.org>


The Linux Scalability project:
	http://www.citi.umich.edu/projects/linux-scalability/




-
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/