Hi guys. I've been following this thread and it occured to me
that it might be possible to write a very quick cache if we
ignored shareability between forked dumps.
I've done so. It actually does appear to make a considerable
difference even with a tiny (4MB default) cache
Matthew Dillon wrote:
> The filesystem wasn't idle and you didn't sync, so of course dump
> screwed up! It would probably screw up on a FreeBSD box too.
What's not idle? The file system is mounted, but there are no writes
to it during the dump. This and other tests run fine in Freebsd
:
:Matthew Dillon wrote:
:> The filesystem wasn't idle and you didn't sync, so of course dump
:> screwed up! It would probably screw up on a FreeBSD box too.
:
:What's not idle? The file system is mounted, but there are no writes
:to it during the dump. This and other tests run fine i
On 2002-Jul-22 12:43:18 -0700, Matthew Dillon <[EMAIL PROTECTED]> wrote:
>Here are the preliminary results when I test this dumping /usr
>to /dev/null:
>
> DUMP: finished in 140 seconds, throughput 6413 KBytes/sec (8 MB cache)
> DUMP: finished in 144 seconds, throughput 6235 KBytes/sec (4 MB cac