Hi
I have just running a couple of benchmarks, with the two last
patchs that I sent to the list:
http://carpanta.dc.fi.udc.es/~quintela/kernel/2.4.0-test8-pre6/ext2_lock_super_05.patch
http://carpanta.dc.fi.udc.es/~quintela/kernel/2.4.0-test8-pre6/ext2_delayed_balance_dirty_03.patch
Without the two patches I obtain the following times running dbench 48
It gives me results in the order:
Throughput 7.24267 MB/sec (NB=9.05333 MB/sec 72.4267 MBit/sec)
real 14m35.883s
user 0m50.570s
sys 1m18.020s
(all between 5-8 MB/sec)
When I applied the two patches, I get a first run very fast:
Throughput 17.0279 MB/sec (NB=21.2848 MB/sec 170.279 MBit/sec)
real 6m13.281s
user 0m50.190s
sys 1m14.700s
Bat the next ones get the same speed than before. This is not only
with test8-pre6, it happens at least form test4 (when I begin to
test), and I haven't seen able to find what is the problem, why I am
not able to get the same speed _always_.
My configurtion is:
SMP Celeron 466
128MB RAM
hda: FUJITSU MPE3084AE, ATA DISK drive
hda: 16514064 sectors (8455 MB) w/512KiB Cache, CHS=1027/255/63, UDMA(33)
(yes the disk is not an _state of the art_ disk, it appears that if
you have faster disks stalls and that problems happen less frequently).
Any comment, sugestion is welcome.
Later, Juan.
--
In theory, practice and theory are the same, but in practice they
are different -- Larry McVoy
-
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to [EMAIL PROTECTED]
