Do larger hard drives last longer?

[army_ant7]

I'm wondering if current generation HDD's are known to have wear-leveling just like SSD's so that if you compare a 1TB to 2TB drive, the 2TB drive theoretically should last twice as long when receiving the same number of writes. Thanks!

 

[apache_lives]

one bad sector can corrupt the entire partition table/boot sector -- technically you can see it as larger the drive the MORE chance of data loss

second larger drives have MORE data to loose, and MORE platters usually too (more heads etc to fail)

just like any storage device BACK UP YOUR DATA DO NOT TRUST OR RELY ON THEM AT ALL

treat them all like there going to fail tomorrow

 

[army_ant7]

Thanks for that perspective on that. I believe it could be quite the opposite with SSD's since they can distribute the writes throughout the drive (wear-leveling), assuming all the sectors were healthy to begin with.

I am guessing as well that you're implying that there isn't wear-leveling on an HDD? I didn't think so as well since with HDD's the outer layers are used for that index (dang it I forgot if it was MBT or what) as well as fragmentation and stuff.

I'm aware about not ever knowing when I drive will fail and the importance of backing up. Thanks for the concern though.

On a side note: I do remember looking up a way to isolate a bad sector (kind of) so that the rest of the drive may live on. Whoever's interested could request for me to dig it up.

 

[4745454b]

HDDs don't need to "wear level" because they don't work like SSDs do. Fragmentation can happen anywhere on a HDD, not just the outer edge. You can mark a bad sector, but with a HDD odds are high that if one starts to go bad, others nearby will start to fail as well.

Some have suggested that newer drives are failing more then the older ones. As they continue to pack bits closer and closer together they seem to be failing more and more. I have yet to see anything scientific confirming this.

 

[army_ant7]

Thanks for the reply! When you say they don't need it, do you mean that they aren't as limited in terms of write cycles as flash memory that everything else would most likely die out (motors, moving parts, etc.) before that happens, or that magnetic disks really don't wear out from re-writing?

Sorry about that, I meant that fragmentation is also an issue if ever there was wear-leveling along with the index on the outer edge and was not just referring to it with the fragmentation.

I have heard of both of those things you said I think, or maybe what I heard with the latter, it was just about if you have a good drive or not 'coz if your drive worked well so far odds are it was manufactured well. I'm kinda unsure if what I read applied to SSD's only or both. Thanks for all that info though.

Along with my question above, does everyone agree that scientifically, larger drives are more prone to failure?

 

[jsrudd]

Larger drives are not more prone to failure. Wear leveling does not apply to HDD.

 

[4745454b]

As far as I'm aware of you can write a bit on a hdd as often as you want. Its not like the NAND found in an SSD.

 

[army_ant7]

It looks like we have conflicting info about how capacity may affect life, unless someone can bring up some hard data. Thanks for sharing that though guys!

I've always thought that writing on a sector on an HDD would wear that sector out. Anyone have info on that?

I really have to think that wear-leveling is kinda impractical on HDD's 'coz of how data is written on them, unless it is totally unnecessary which I'm curious why if ever.

 

[RetiredChief]

SSD have a built in failure, There is a limited number of writes that a NAND cell can sustain. HDDs do not have this inherant failure as the magnetic domain can have its polarity reverse as long as it remains intact.

An SSD uses an electrical grid axis (X,Y,Z being the charge) this axis is not mecahanically controlled. HDDs also X,Y axis (Z being the magnectic polarity), But must be access by mechanical means. mechanically. For HDDs the ability to maintain the sector alignment increase is more difficault as the domains become smaller.

But to the question which was do Larger HDDs have a higher failure rate. I've not seen a report that shows this, But if you look at new egg, you could get the impression that as the size goes up from 1 TB the problems seems to also increasa, Much more so for the Consummer class HDDs than the enterprise class drives (But then they cost alot more.

There is another problem in that there is a specified bit error rate (or some such verbage). With the old smaller HDDs it was just a number eggheads tossed around and totally exceeded the size of HDDs. There were some articals a couple of years ago on this as HDD size was expected to exceed this magical number - and that was before the 1 TB drives - Have not seen anything more on this.

Added - there is NO wear leveling algortium used on a HDD as there is no cell (domain) degregation based on the Number of writes.
What HDDs do imploy is added bits in a sector that contains alignment info. If the head can not read the alignment data the sector is "lost" and in many cases is why the data can not be read from a given sector, It's not the domain is BAD, it just "lost" the location.

 

[army_ant7]

Best answer selected by army_ant7.

 

[army_ant7]

I guess I got my explanation/answer. Thanks everyone, especially RetiredChief for being really thorough! I was able to get it. :-D