2.5" Reliability, compared to 3.5" drives?

Silencing hard drives, optical drives and other storage devices

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
intx
Posts: 175
Joined: Fri Apr 23, 2004 12:57 am
Location: Ontario, Canada

2.5" Reliability, compared to 3.5" drives?

Post by intx » Fri Jul 21, 2006 4:29 pm

Between 2.5" HDs and 3.5" HDs, which are more reliable? I'd think in theory 2.5" HDs are more reliable because they generate less heat, which seems to be the leading factor determining the life your HD.

However, they are smaller, and supposedly built more fragile.. does this equal less reliability? The only comments on this forum seem to learn towards 3.5" HDs being more reliable.

On the otherhand, that misconception could simply be because most 2.5" HDs are placed in laptop/notebooks, which place the HDs in a more prone-to-failure environment (more heat, higher number of start/stops, shock, etc.)

I've search the forums, but there doesn't appear to be a discussion on this. I'm sorry if this is a repost.

VERiON
Posts: 233
Joined: Wed Aug 11, 2004 5:42 am
Location: EU

Post by VERiON » Wed Jul 26, 2006 3:31 am

I've tried to start disscussion about 2,5" hdds reliability in this thread.

My post didn't bring much attention - no answers at all.
But you can find some interesting links in my post - mostly about MTBF and disc reliability in general.

In short:
2,5" = more start/stop cycles, less MTBF hours
3,5" = less start/stop cycles, more MTBF hours

jaganath
Posts: 5085
Joined: Tue Sep 20, 2005 6:55 am
Location: UK

Post by jaganath » Wed Jul 26, 2006 4:24 am

However, they are smaller, and supposedly built more fragile
Why would 2.5" be more fragile than their 3.5" counterparts? One would assume that 2.5" HDD are built more robustly, to withstand higher operating shock, temperature and consume less energy, in order to cope with laptop applications; therefore 2.5" when used in desktop should be more durable than 3.5", due to more robust design, and also less mechanical stress, lower platter RPM, etc. Many SPCRer's use 2.5" in desktops, I don't recall anyone reporting a HDD failure (even though many use enclosures which raise temps).

CoolGav
Posts: 398
Joined: Wed Jul 23, 2003 4:26 am
Location: €ngland
Contact:

Post by CoolGav » Wed Jul 26, 2006 8:43 am

I've recently experienced problems on a Samsung 2.5" 80GB drive (MP0804H I think). I got 2 just over a year ago. 1 went into a laptop that doesn;t get used a lot, anopther into a desktop that also doesn't get used that much. The desktop drive was not activly cooled (neither was the latop, but it was surrounded by metal in the latop), just suspended where there should have been some airflow.

The problem isn't that it's died, but that it keeps losing important Windows files, meaning it stops the PC from booting. So some corruption. I repartitioned and changed to an Antec Solo with active cooling for it, and it's done it again after at least 10 days being OK. I'm giving up on that drive for an OS, and will see how it does with just data (stuff I could lose I guess!).

I wouldn't go so far to say laptop drives are less reliable, but I wonder...

BTW, I've got a WD drive on order to replace it!

arrikhan
Posts: 79
Joined: Thu Dec 29, 2005 3:51 am
Location: Australia

Post by arrikhan » Thu Jul 27, 2006 4:50 am

Where I work, we install commodity intel servers like they are going out of fashion (~70/month) and we've just moved from dell dual cpu and quad cpu boxes with 3.5" drives to IBM dual cpu and quad cpu (dual core) intel servers as our standard with 2.5" drives.

I'm interested to know if there is a different reliability in these drives!

I would have thought the form factor (MB's per inch) was the same, therefor less energy is being expended to spin less metal, thus less heat. Based on the same method to remove heat (attached to frame and a few fans) everything else is pretty much the same...

The question is, is the motor the same or different? If it's the same, it's doing less work thus less likely to fail as it isn't working 'as hard'. If it's different, we're into the 'depends' response based on reliability of newer technology of a newer, maybe smaller motor component.

Arrikhan

Post Reply