Re: hard- or software-raid?
On Fri, 2003-01-24 at 09:22, Adrian 'Dagurashibanipal' von Bidder wrote:
> On Fre, 2003-01-24 at 00:16, Tinus Nijmeijers wrote:
> > I'm building a server that needs about 200G of harddisk space and the
> > data has to be safe. If I need to replace a faulty hd and get downtime
> > that's fine. Speed is not an issue.
>
> Agree with Russel and thing: what's raid6?
>
my other system:
***** DAC960 RAID Driver Version 2.2.10 of 1 February 2001 *****
Copyright 1998-2001 by Leonard N. Zubkoff <lnz@dandelion.com>
Configuring Mylex AcceleRAID 170 PCI RAID Controller
Firmware Version: 6.00-01, Channels: 1, Memory Size: 32MB
PCI Bus: 0, Device: 11, Function: 1, I/O Address: Unassigned
PCI Address: 0xFB000000 mapped at 0xFC800000, IRQ Channel: 17
Controller Queue Depth: 512, Maximum Blocks per Command: 2048
Driver Queue Depth: 511, Scatter/Gather Limit: 128 of 257 Segments
Physical Devices:
0:2 Vendor: FUJITSU Model: MAN3184MC Revision: 0107
Wide Synchronous at 160 MB/sec
Serial Number: UFL0P15005AC
Disk Status: Standby, 35852288 blocks
0:3 Vendor: FUJITSU Model: MAN3184MC Revision: 0107
Wide Synchronous at 160 MB/sec
Serial Number: UFL0P150058H
Disk Status: Online, 35852288 blocks
0:4 Vendor: FUJITSU Model: MAN3184MC Revision: 0107
Wide Synchronous at 160 MB/sec
Serial Number: UFL0P15005TV
Disk Status: Online, 35852288 blocks
0:5 Vendor: FUJITSU Model: MAN3184MC Revision: 0107
Wide Synchronous at 160 MB/sec
Serial Number: UFL0P15005SS
Disk Status: Online, 35852288 blocks
0:6 Vendor: FUJITSU Model: MAN3184MC Revision: 0107
Wide Synchronous at 160 MB/sec
Serial Number: UFL0P15006BD
Disk Status: Online, 35852288 blocks
0:7 Vendor: MYLEX Model: AcceleRAID 170 Revision: 0600
Wide Synchronous at 160 MB/sec
Serial Number:
Logical Drives:
/dev/rd/c0d0: RAID-6, Online, 71704576 blocks
Logical Device Initialized, BIOS Geometry: 255/63
Stripe Size: 64KB, Segment Size: 8KB
Read Cache Disabled, Write Cache Disabled
No Rebuild or Consistency Check in Progress
RAID 6
RAID 6 is similar to RAID 5, but uses two parity schemes simultaneously.
This gives it the advantage of very high fault tolerance, but at the
cost of write performance (as there is a lot of parity information that
needs to be computed and written), and a relatively high overhead.
tinus
Reply to: