bug-hurd
[Top][All Lists]
Advanced

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

Re: removing an ext2fs file forces disk activity


From: Niels Möller
Subject: Re: removing an ext2fs file forces disk activity
Date: 04 May 2002 22:41:36 +0200
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2

tb@becket.net (Thomas Bushnell, BSG) writes:

> nisse@lysator.liu.se (Niels Möller) writes:
> 
> > Does anybody have any idea how often this case occurs with typical
> > activities like compilation?
> 
> What's the point of the question: to decide if we can ignore the
> issue, or to decide if the solution has to be terribly efficient?

I'd like to know if implementing the new scheme would really help
making the statement "The Hurd is significantly slower than Linux for
things like big compiles" false. If it's an optimization that's worth
the effort.

As for "terribly efficient", I'm simply assuming that the cost will be
that one sync of the intermediate state of A that is required for
robustness, that the book-keeping is free, and that it's quite hard or
complicated (say, introducing some filesystem transaction log) to
improve on that.

It seems relevant to ask if the typical file access patterns will
generate mostly acyclical dependencies between blocks (in which case
the new scheme is a big win, as no waiting for blocks to be synced is
needed at all, if I understood it correctly). Or if the A, B, A
pattern is typical, say it happens about ones every time a file is
created or modified or something like that, in which case we would
still a lot of syncing.

Best regards,
/Niels



reply via email to

[Prev in Thread] Current Thread [Next in Thread]