On 14. jan.. 2009, at 12.17, Jasse Jansson wrote:
First of all, thanks Simon for the links, It sure was interesting
reading.
On Jan 14, 2009, at 11:45 AM, Bjørn Vermo wrote:
I have toyed with the idea that one drive in a mirrored pair ought
to have the addresses inverted, so sector 1 on one drive is mapped
to sector MAX -1 on the other.
I read it like the locality (radius) of the errors occur on the same
disc,
so I don't see the benefit from your "reversed mirror strategy".
He clearly stated that some drives models had error spikes at certain
cluster numbers.
This should not really surprise anybody who is familiar with the way
modern drives work internally. The cylinder where you change to a
different data density is a border case, and any border cases open
extra bug opportunities.
You will also get some interesting performance problems to tackle,
when
one disc works on the inner tracks and the other one is on the outer
tracks.
A smart controller could make that into an opportunity. It is hardly
any more difficult than staggered stripes, which is widely implemented.
In most mirroring controllers I have seen the system will get the read
data from the first drive that has them ready. Two "opposite" drives
should be able to offer a better average read performance.
--
Bjørn Vermo
Core networking
Opera Software ASA
