Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 07 Apr 2008 20:48:05 +0200
From:      Ferruccio Zamuner <nonsolosoft@diff.org>
Cc:        freebsd-geom@freebsd.org
Subject:   Re: gmirror on a very slow but new HP architecture
Message-ID:  <47FA6C65.1010007@diff.org>
In-Reply-To: <47F9489C.7040602@quip.cz>
References:  <47F92BD9.1060207@diff.org> <47F9489C.7040602@quip.cz>

next in thread | previous in thread | raw e-mail | index | archive | help
Miroslav Lachman ha scritto:
> Did you tried the same dd test with another OS (with some bootable 
> live CD of Linux)?
>
> It may be caused by bad disk controller / bad cables.
>
> Miroslav Lachman
With Ubuntu kernel 2.6.20 it transfer 6-7 Mb/s, still very slow for a sata.

Then I've drop the HP LSI Logic SATA/RAID controller and I've connected 
hds on the mainboard controller
and on a Promise TX2 SATA300 that finally it's supported by FreeBSD 7.


Results are better but still non good as they should be:

free# gmirror list
Geom name: gm0
State: COMPLETE
Components: 3
Balance: split
Slice: 8192
Flags: NONE
GenID: 0
SyncID: 2
ID: 3991481480
Providers:
1. Name: mirror/gm0
   Mediasize: 237291069952 (221G)
   Sectorsize: 512
   Mode: r6w6e6
Consumers:
1. Name: ad6s1
   Mediasize: 237291070464 (221G)
   Sectorsize: 512
   Mode: r1w1e1
   State: ACTIVE
   Priority: 0
   Flags: NONE
   GenID: 0
   SyncID: 2
   ID: 3276824044
2. Name: ad10s1
   Mediasize: 237291070464 (221G)
   Sectorsize: 512
   Mode: r1w1e1
   State: ACTIVE
   Priority: 0
   Flags: NONE
   GenID: 0
   SyncID: 2
   ID: 1289844737
3. Name: ad12s1
   Mediasize: 237291070464 (221G)
   Sectorsize: 512
   Mode: r1w1e1
   State: ACTIVE
   Priority: 0
   Flags: NONE
   GenID: 0
   SyncID: 2
   ID: 1706660870

free# swapoff /dev/mirror/gm0b
free# dd if=/dev/zero of=/dev/mirror/gm0b bs=8192 count=10000
10000+0 records in
10000+0 records out
81920000 bytes transferred in 2.443927 secs (33519822 bytes/sec)
free# dd if=/dev/mirror/gm0b of=/dev/null bs=8192 count=20000
20000+0 records in
20000+0 records out
163840000 bytes transferred in 7.773088 secs (21077852 bytes/sec)
free# dd if=/dev/mirror/gm0b of=/dev/null bs=8192 count=20000
20000+0 records in
20000+0 records out
163840000 bytes transferred in 7.444171 secs (22009167 bytes/sec)
free# dd if=/dev/zero of=/dev/mirror/gm0b bs=8192 count=20000
20000+0 records in
20000+0 records out
163840000 bytes transferred in 5.668771 secs (28902208 bytes/sec)
free# dd if=/dev/mirror/gm0b of=/dev/null bs=8192 count=20000
20000+0 records in
20000+0 records out
163840000 bytes transferred in 8.050144 secs (20352430 bytes/sec)


Why are there so poor performances?

Any idea?

More info from dmesg:

atapci0: <Promise PDC40775 SATA300 controller> port 
0x4400-0x447f,0x4000-0x40ff mem 0xefa20000-0xefa20fff,0xefa00000-0xefa
1ffff irq 18 at device 1.0 on pci10
atapci0: [ITHREAD]
atapci0: [ITHREAD]
ata2: <ATA channel 0> on atapci0
ata2: [ITHREAD]
ata3: <ATA channel 1> on atapci0
ata3: [ITHREAD]
ata4: <ATA channel 2> on atapci0
ata4: [ITHREAD]
[..]
atapci2: <Intel ICH7 SATA300 controller> port 
0x30c8-0x30cf,0x30bc-0x30bf,0x30c0-0x30c7,0x30b8-0x30bb,0x3090-0x309f 
mem 0xefd00400-0xefd007ff irq 19 at device 31.2 on pci0
atapci2: [ITHREAD]
ata5: <ATA channel 0> on atapci2
ata5: [ITHREAD]
ata6: <ATA channel 1> on atapci2
ata6: [ITHREAD]
[..]
ad6: 238475MB <WDC WD2500YS-70SHB1 20.06C06> at ata3-master SATA150
ad10: 238475MB <WDC WD2500YS-70SHB1 20.06C06> at ata5-master SATA150
ad12: 238475MB <WDC WD2500YS-70SHB1 20.06C06> at ata6-master SATA150




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?47FA6C65.1010007>