Tugger the SLUGger!SLUG Mailing List Archives

Re: [SLUG] LVM


On 15/06/10 17:10, james wrote:
On Tuesday 15 June 2010 12:15:52 you wrote:
On Tue, Jun 15, 2010 at 11:14:38AM +0800, james wrote:
The stuff  below is interesting and a reference, but this highlights my
favourite rant: Seagate's 'ATA more than an interface' says multiple
disks in a machine *will* result in a higher failure rate, maybe much
higher.
Due to heat, or what? That paper seems to concern itself primarily with the
differences between PS (personal storage) drives and ES (enterprise
  storage), in order to justify why the SCSI drives have so much higher cost
  per bit.
The bit that says:
Disk#1 seeks knocking Disk#2, Disk#3 off track
so
Disk#2 seeks knocking (mechanical coupling) Disk#1 off track
so
Disk#1 seeks again
etc

my own experience is that n-disk arrays fail more than n times 1 disk
but that is oh so subjective, and so subject to the ravages of stats.

James
in subjective land disks in arrays are probably used allot more heavily than those sitting in singles (IE server Vs Desktop usage)

I understand that server grade disks have accelerometers built in these days to pick up accelerations caused by other disks in the stack seeking.

I also recall seeing a video of a really large array (40+ disks in a dedicated storage box) in a data centre, they had pulled up a latency monitoring tool and then had a guy scream at the drives, this caused a noticeable spike in latency