[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: journal writeback interval (Re: how to counteract slowdown)

On Mon, 2001-11-12 at 22:31, Jeffrey W. Baker wrote:

> Obviously, there were a lot of missed opportunities to write some blocks
> out in those blank spaces.  I'd much rather see 500 blocks out per
> second than 2500 blocks out in one second, every five seconds.  

If you were to run your test several times, mounting as ext2 a few times
and ext3 a few times (and either unmounting or rebooting in between
runs, to clear the caches), *and* you provided both overall wallclock
times as well as vmstat output, then it mgiht be possible to draw some
meaningful conclusions.

In the absence of bags and bags of comparative measurements, all we have
is handwaving and mushy notions, which as the 2.4 VM saga demonstrates,
doesn't get us very far.

> [2] An aside: it seems that an 'updatedb' program which knew a lot about
> ext2's structure could be smarter.

Perhaps, but it would also be vastly more complex, and it would
assuredly not work well for any other kind of filesystem.  A poor
tradeoff for a batch job that runs at 3am.


Attachment: pgp00005.pgp
Description: PGP signature

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]