[geeks] Fwd: [IP] Interesting speculation on the tech behind gmail

[Charles Shannon Hendrix]

Wed, 07 Apr 2004 @ 18:14 +0100, Mike Meredith said:

> On Wed, 7 Apr 2004 08:22:46 -0400 (EDT), Sandwich Maker wrote:
> > "> The most obvious challenge is the storage. You can't lose people's 
> > "> email, and you don't want to ever be down, so data has to be 
> > "> replicated. RAID is no good; when a disk fails, a human needs to 
> > "> replace the bad disk, or there is risk of data loss if more disks 
> > "> fail. One imagines the old ENIAC technician running up and down the
> > 
> > this is so glibly wrong.
> 
> Well except in one bit accidentally ... with a large enough number of
> disks you will need someone running around replacing the failed ones on
> a full-time basis.

I'm not so sure.

Drive density is tremendous now, and a petabyte fits in a smaller space
than it used to.  With hot spares and maybe also redundancy among RAIDs,
you should only have to do drive replacement at scheduled intervals.

Large IBM mainframes were using robots for drive media over 20 years
ago.  I would not be at all surprised to see really huge RAID farms do
something like that.  Certainly at some point it becomes cost effective.

Of course, hopefully they will work better than some of the tape robots
I've had to use, which threw tapes on the floor all the time.

-- 
shannon "AT" widomaker.com -- ["I want this Perl software checked for
viruses.  Use Norton Antivirus." -- Charlie Kirkpatrick]