Andrew Lentvorski wrote:
> Tracy R Reed wrote:
> 
>> - - SCSI, FC, SATA, ATA are equally reliable (surprising and embarrassing
>> to people who have spent big bucks on SCSI/FC but both studies came to
>> the same conclusion based on 100,000 disks each)
> 
> While I'm certainly someone who advocates SATA in RAID 1, I'm
> unconvinced about the comparisons with SCSI.
> 
> I don't recall any of the studies actually using 10K SATA drives and
> comparing them to 10K SCSI drives.  And, for 15K drives, the only option
> is still SCSI.
> 
> That means that they are likely comparing fairly new, large ATA 7200RPM
> drives to *significantly* older 7200RPM SCSI drives.  That's a bit unfair.
> 
>> Unfortunately google wussed out and won't tell us whose drives are the
>> most/least reliable. The second study didn't mention this either. I
>> guess they are afraid of getting sued.
> 
> Actually, I found the Google paper a prime example of academics creating
> a not terribly useful paper because it lacks details.  They hide the raw
> data, it is difficult to reproduce, and it lacks predictive power.
> 
> We would be better off if they published the raw data without comment.

I seem to be more willing to accept their paper as being a reasoned
interpretation of what must be an enormous amount of raw data.

Although, they admittedly do it through handwaving discussion (of a few
sub-sets of statistics), I got the impression they took some care to
remove the influence of wildly different hardware types from their
general conclusions.

I took (assumed) their working under the Google name to be a hint they
would likely be more interested in real-world conclusions (ie, of value
to their operations), than in the ivory-tower world .. but who knows?

Regards,
..jim


-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to