Re: Some very thought-provoking ideas about OS architecture.

Malcolm Beattie (mbeattie@sable.ox.ac.uk)
Mon, 21 Jun 1999 15:55:51 +0100 (BST)

Pavel Machek writes:
> > I don't think it is really so important. If system crashes, something is
> > bad anyway, and you should cure the causation (fix bugs), not the
> > consequence (do quick recovery).
> > Power fault is another case, but I think it really doesn't matter if
> > system is down for 3 hours because of power fault or 3 hours and 10
> > minutes because of power fault and fsck.
>
> Or 3 hours of power fault and 3 hours of fsck... Which is exactly what
> happens on ~100Gig disks.

Please don't guess/exaggerate: I've already posted fsck benchmark
times to this mailing list a month ago. In particular, fsck for a
43 GB ext2 filesystem (4K blocks) with 30 GB in use took 13 minutes.

Disk usage was 25 directories each with a copy of the Linux 2.2.1
source tree (63MB, 4000 files) and each with 200 subdirectories
holding 5 files of 1MB.

That was with a single SCSI bus connected to a (software) RAID5 array
of 6 x 9GB (10K RPM) disks. Triple that to a three-way SCSI adapter
connected to one 43 GB array each and fscking the whole damn lot will
run in parallel and take about the same time: 13 minutes. Not 3 hours.
The person you replied to who said an additional 10 minutes for fsck
was far closer then your guess.

If you want to treat 100 GB of disk like tape with slow access then
fsck will take longer but any decently configured system with a lot of
disk should also have a decent I/O subsystem.

--Malcolm

-- 
Malcolm Beattie <mbeattie@sable.ox.ac.uk>
Unix Systems Programmer
Oxford University Computing Services


-
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/