Not Much Happening in Hard Drives This Year
yahooooo writes "CoolTechZone.com has an article that talks about desktop hard drive developments in 2005. It looks this year is going to be a dud for the storage industry."
← Back to Stories (view on slashdot.org)
there is not much demand for higher capacities (very few people would need >160gb).
as for reliability, most HD's are acceptable, but you can never fully rely on them to never fail, you must always have a backup system for important data.
speed is one of the areas which is always welcome for improvement (until of course it reaches the max interface speed, eg 150mB/sec for SATA)
Marge, get me your address book, 4 beers, and my conversation hat.
Most of the best news for most HD consumers is price drops, which will probably accelerate. Most of the HD price reflects recouping investment in R&D and retooling factories, not a per-unit cost. So HD companies aren't spending lots more money this year - that means they'll be charging even less, competing on price without other differentiators.
For consumers, that can mean qualitative improvements through passing quantitative thresholds. Buy 2 HDs instead of 1, make a RAID, and watch both uptime and fault recovery become minor bumps in the road, rather than a job-threatening days-long surprise nightmare. While filling the coffers of the vendors, who can reinvest in integrating that kind of redundancy in the HD unit itself. This year's nonevents might just give sysadmins the chance to become the most obviously important link in the IT chain, eclipsing the usually exaggerated developer rockstars.
FWIW, HD consumers probably aren't defined by "HDs", but rather storage in any medium, determined by usage. So the real news in "HD" is really Flash memory, which is seeing huge leaps in capacity, cheapness, perfomance and manageability. When will someone ship a $100 SDIO 1GB/WiFi card? With gumpack-sized, 8-SDIO-socketed battery for a pocket-PSAN (Personal Storage Area Network)? Or start sewing these things into hats and sweatjackets?
--
make install -not war
Thats what I'm waiting for.
I have 3x200gb, 2x160gb, 2x120gb, 4x80gb (and more down the line).
The 200gbs are running at 83% full because... they all mirror each other.
Yup I know it's particularly anal, but I'll agree with the first post: We need more reliable drives. All of my photos are backed up 2x on DVD- one goes into a jukebox, the other goes onto a spindle, and all are stuffed into something called CDStorageMaster (fun proggy).
The HDs mirror each other but I've not yet had time to test a catastrophic failure of this. I had a manual raid before and, when my system crashed due to a bad PSU (note: Antec replaced it free of charge) I was eventually able to get all the drives back up and running, but I was left with a very nasty taste of bad-dynamic disks in my mouth.
So please... more storage at 35cents/gb and I'll be happy. Or 3.5 cents/gb would make me happier, but one can hope.
Four hard drives in four years? I'll admit modern disks are built poorly, but that seems excessive. It's possible that you have had a string of bad luck, but if they all failed in the same machine, you might want to check you power supply and/or cooling setup. The drives might have been killed.
"God is a comedian playing to an audience too afraid to laugh." -Voltaire
I run this test on EVERY new/used hard drive I buy, before putting it into regular use:
/dev/hdX ' /dev/hdX '
' hdparm -c1 -d1
' time badblocks -c 256 -n -s -v
--Using this method on newly-delivered HDs has allowed me to RMA them right away, before they fail with MY data on them.
.
== WolfriderV6 == I'm willing to admit that *I just might* be wrong... Are you??
I want a drive with more features for notebooks...
Sure 4200rpm may save battery life, but they're so god aweful slow. Why don't they make a drive that has variable rpm? You could even have the OS control the speed: 4200 when on battery and 7200 when plugged into an outlet. Maybe even have an override so you can make it fast at the expense of battery life, should you want to.
Lets take a break for the quest to be first with a small
form factor terabyte drive. Instead lets concentrate on
two things:
a) faster. much faster
b) self mirroring (ie raid 1) drives in the same form
factor.
The first is obviously a desire everybody wants.
The second is similar I guess to dual core cpu's vs
dual cpu's. Take a drive and instead of making it 500GB
give me 2 200GB drives on seperate controllers and power
supplies with an internal interface that allows one to
mirror the other. Seemlessly.
While fault tolerance should never be confused with a
'backup', something like this would be very useful. With
giant capacities now prevalent, most consumers have given
up on backing up. But by offering a self contained
fault tolerance you allow the consumer to easily chose
between giant capacity or smaller size but some safety
built in.
For the performance crowd, many who now use raid 10 arrays,
you cut the drive clutter in half. Two bays, not 4 (or 4
not 8). Perhaps you could even get better thermal
peformance than 2 independent drives.