Question Is 2TB, 4TB or 5TB portable hard drive more durable?

[brannsiu]

I remember there was some saying that some certain capacities are more durable than another, is it real?

 

[Lutfij]

Not really, all mechanical drives should be treated with care and caution especially when you factor in the amount of data larger drives can and will hold. Some brands might get favored over another aside from capacities but for me, I tend to stay away from WD drives unless they are the only brand with the capacity I'm looking for and if that's the only brand left when I'm out to purchase said mechanical HDD.

 

[SkyNetRising]

I remember there was some saying that some certain capacities are more durable than another, is it real?

If those portable drives are SSD, then they will be more durable.
Mechanical drives rarely survive being dropped. SSD will not care about that.

 

[ErickParker]

In principle yes, as well as some brands have less problem rates.
2TB and 4TB HDDs are the most durable, while 3TB has a history of issues. Generally speaking, you can rely on your hard drive for three to five years on "average".
This statement comes from Backblaze, the online backup company who analyzed the failure rates of 25,000 running hard drives.
They found that 90% of hard drives survive for three years and 80% for four years. But this number varied across brands. Western Digital and Hitachi hard drives lasted much longer than Seagate's in Backblaze's study.
The test shows that 5.1% fail in the first year due to manufacturing problems! Between the second and third years, only 1.4 fail and in the fourth year the percentage of failure rises to 11.8%.
Even avoiding moisture, dust, shocks, variations in electrical current, HDDs tend to fail due to wear and tear on moving components.
Also consider that a 5400rpm HDD will last longer than a 7200rpm one because wear and noise will be less.
My recommendation would be WD Blue 2TB or 4TB 5400rpm which have a good reliability/price ratio

 

[USAFRet]

I remember there was some saying that some certain capacities are more durable than another, is it real?

For spinning HDD, no it is not.

 

[USAFRet]

In principle yes, as well as some brands have less problem rates.
2TB and 4TB HDDs are the most durable, while 3TB has a history of issues.

There was a short term spate of a particular model of 3TB Seagate that had issues.

Looking at BackBlaze results over the last few years, there is little difference across brands or models.

For 2021, 2% AFR or less.

Backblaze Drive Stats for 2021

Understanding our hard drives is key to understanding our company. Read the Backblaze drive stats for 2021 to learn more about our expansions.

www.backblaze.com

For instance...
The 16TB Toshiba Enterprise, model num MG08ACA16TE has an annual fail rate of 0.91%
Pretty good, right?

However, the one and only one I had failed within 7 months.
Its replacement was delivered 2 weeks ago, and is sitting in the dock next to me.

Fail rates like that only count on a fleetwide basis, for budgeting and mass buys.
The particular drive YOU have may fail, no matter who says what.

 

[fzabkar]

Seagate and WD both produce "flimsy" 2.5" external drives. These designs are constrained by height ("slim" models) and the fact that the platters are thinner and heads are smaller and more fragile.

Personally I would opt for a 3.5" drive in a well ventilated metal case, preferably with a fan. Never mount these drives vertically, even if they are designed that way. If they fall over, bye bye data.

 

[cryoburner]

If you are primarily concerned about durability to avoid the loss of irreplaceable data, your best option is to have all important data stored on at least two drives. That way, if one drive fails, the data is likely to still be safe on the other. Just be sure to replace the failed drive with another as soon as possible.

This statement comes from Backblaze, the online backup company who analyzed the failure rates of 25,000 running hard drives.

Keep in mind, Backblaze is company that's using consumer hard drives in a commercial, server-type setting, so the operating conditions are going to be different from what a drive is likely to experience in a typical system. They may see higher temperatures, and likely increased vibration, as they mount up to 60 drives together in a single case, with 20 of those cases stacked together in each server rack. Scroll down a bit here to see how the drives are mounted in their current "pod" design...
https://www.backblaze.com/blog/open-source-data-storage-server

As a result, their failure rates may not necessarily reflect the typical failure rate of a given drive (though could potentially indicate if some design might have a problem). There's also the possibility that their sample of drives may or may not happen to include a bunch of drives from a "bad batch", so while their results might be interesting, you should should take them with a grain of salt.

And of course, these are all full-size desktop hard drives, and if someone is looking at smaller "portable" external drives, those will be using laptop drives inside, with entirely different platters from the ones Backblaze has in their systems.

 

[Banqu0]

As some have pointed out, reliability issues for specific capacities are mostly brand specific. Though it's fair to point out that newer products are always going to be riskier from a consumer standpoint because they haven't been time tested. This doesn't necessarily mean that the newest high capacity drive should be avoided, but it pays to research the specific drive you intend to purchase. Data is pretty valuable.

That said, 3.5 inch drives are going to be more robust than the 2.5 inch ones.

 

[USAFRet]

If I were to research ALL drives in the capacity I wanted, from multiple reputable sources...and found some particular model with 0.0% fail....and buy it.

I would STILL bring it into my current backup routine.

Data that exists on a single device can be considered to not exist at all.

Trust, but verify

 

[cryoburner]

That said, 3.5 inch drives are going to be more robust than the 2.5 inch ones.

Do you have actual evidence for that though? It's possible, though I would imagine that how the drives are typically used plays much more of a factor.

2.5" drives will mostly be installed in laptops and portable external drives, making them more likely to get transported and jostled around on a regular basis, in addition to seeing higher temperatures due to lack of airflow.

3.5" drives, on the other hand, will mostly be installed in things like desktop computers and servers, which should remain mostly stationary and usually have better cooling. Even 3.5" external drives are more likely to be left stationary most of the time, and won't be as likely to get thrown into a backpack or something.

So even if there is some evidence in terms of product failure rates, it might be mostly down to how the drives are typically handled more than anything.

 

[ErickParker]

If you are primarily concerned about durability to avoid the loss of irreplaceable data, your best option is to have all important data stored on at least two drives. That way, if one drive fails, the data is likely to still be safe on the other. Just be sure to replace the failed drive with another as soon as possible.


Keep in mind, Backblaze is company that's using consumer hard drives in a commercial, server-type setting, so the operating conditions are going to be different from what a drive is likely to experience in a typical system. They may see higher temperatures, and likely increased vibration, as they mount up to 60 drives together in a single case, with 20 of those cases stacked together in each server rack. Scroll down a bit here to see how the drives are mounted in their current "pod" design...
https://www.backblaze.com/blog/open-source-data-storage-server

As a result, their failure rates may not necessarily reflect the typical failure rate of a given drive (though could potentially indicate if some design might have a problem). There's also the possibility that their sample of drives may or may not happen to include a bunch of drives from a "bad batch", so while their results might be interesting, you should should take them with a grain of salt.

And of course, these are all full-size desktop hard drives, and if someone is looking at smaller "portable" external drives, those will be using laptop drives inside, with entirely different platters from the ones Backblaze has in their systems.

Thank you for the link. Very interesting.
I understand your point of view, but in real life, computers are not in a clean, dust-free environment with constant temperature 24/7.
So I think the test with 25,000 HDDs is valuable and useful for a consumer's decision making.
We have a small design office and in the last 3 years only Seagate HDDs have failed. Coincidence ? Bad luck ? I don't know, but all have been replaced by HDD from Toshiba and WD or SSDs from Crucial.

 

[ErickParker]

There was a short term spate of a particular model of 3TB Seagate that had issues.

Looking at BackBlaze results over the last few years, there is little difference across brands or models.

For 2021, 2% AFR or less.

Backblaze Drive Stats for 2021

Understanding our hard drives is key to understanding our company. Read the Backblaze drive stats for 2021 to learn more about our expansions.

www.backblaze.com

For instance...
The 16TB Toshiba Enterprise, model num MG08ACA16TE has an annual fail rate of 0.91%
Pretty good, right?

However, the one and only one I had failed within 7 months.
Its replacement was delivered 2 weeks ago, and is sitting in the dock next to me.

Fail rates like that only count on a fleetwide basis, for budgeting and mass buys.
The particular drive YOU have may fail, no matter who says what.

Thank you for the link. I read the new data carefully and noticed that in the 2021 survey there are few WD HDDs.

Although the failure rates are really close, as you said, on average the Seagates are still the ones that fail the most (except for this 6TB model) and the ones with the lowest failure rates are still the HGST.
However, if you are a person pursued by bad luck, it doesn't matter which brand you choose

 

[USAFRet]

We have a small design office and in the last 3 years only Seagate HDDs have failed. Coincidence ? Bad luck ? I don't know, but all have been replaced by HDD from Toshiba and WD or SSDs from Crucial.

And my last two dead HDD was 3TB WD and 16TB Toshiba.
So.....

 

[smalltech]

In principle yes, as well as some brands have less problem rates.
2TB and 4TB HDDs are the most durable, while 3TB has a history of issues. Generally speaking, you can rely on your hard drive for three to five years on "average".
This statement comes from Backblaze, the online backup company who analyzed the failure rates of 25,000 running hard drives.
They found that 90% of hard drives survive for three years and 80% for four years. But this number varied across brands. Western Digital and Hitachi hard drives lasted much longer than Seagate's in Backblaze's study.
The test shows that 5.1% fail in the first year due to manufacturing problems! Between the second and third years, only 1.4 fail and in the fourth year the percentage of failure rises to 11.8%.
Even avoiding moisture, dust, shocks, variations in electrical current, HDDs tend to fail due to wear and tear on moving components.
Also consider that a 5400rpm HDD will last longer than a 7200rpm one because wear and noise will be less.
My recommendation would be WD Blue 2TB or 4TB 5400rpm which have a good reliability/price ratio

I am going to buy a Seagate 2.5inch external hard drive either 4TB or 5TB. May I know is 4TB or 5TB better?

I am not sure if all odd numbers TB have higher failure rates or not.

Side note I saw this report (not sure how to read it) and all the TB listed in the chart are even number TB, why no odd number TB? https://www.backblaze.com/blog/backblaze-drive-stats-for-2021/

 

[mdd1963]

No one is concerned over even/odd numbers in a drive's capacity. (I had heard years back that a few manufacturers' 3 TB models sucked/tended to fail sooner than smaller or larger drives, but, I've had a 3 TB Toshiba external for 5 years that is working fine, but, as it's only connected/powered on maybe 4 -6 hours per year anyway, I'd not make any 'what great reliability' conclusions based on that...)

Get whichever drive is priced better per TB...(with most external 3,4-5 TB external USB drives on sale for sub-$90 these days, this is not quite like choosing between expensive sports cars...

 

[USAFRet]

I had heard years back that a few manufacturers' 3 TB models sucked/tended to fail

Yes, there was a particular run of Seagate 3TB drives that were problematic.
But that was very much an outlier.

Overall, there isn't more than a percentage point or two difference between manufacturers and sizes.
And that is only on a fleetwide basis.