Large Capacity HDD's (6TB-20TB)

badwhitevision said:
OT, but could you please point me to a few sources or explain why?

My reddit searches only talk about the end user being stupid or careless and not about the interface itself. I'm currently running such a setup and hence wondering.

Thanks.
Click to expand...
there are many, it's common knowledge that the USB interface is not as reliable as SATA or NVME.
Source 1: https://www.truenas.com/community/threads/external-usb-bays-a-good-idea-or-a-bad-idea.116403/
Source 2: ChatGPT
1720097795125.png

Source 3: https://forums.truenas.com/t/why-you-should-avoid-usb-attached-drives-for-data-pool-disks/1499
TLDR If you are using the enclosure and HDD inside it as just a disk with data i.e. no RAID of any kind, and manually backing it over to another disk. You should be fine with using USB Enclosures for storing your data and making it available on the network. but if you want to do a RAID or ZFS or whatever anything that requires maintaining parity on multiple drives then you should be looking at a NAS or DIY solution and not USB enclosure.
 
Can confirm that USB enclosures can be unreliable. The drives I got were incorrectly showing SMART data issues when checked with an Orico enclosure. But I have them plugged in directly using SATA for around two weeks now and have had no issues so far.
 
If anyone is concerned about the Reliability Factor and wants to check out the Failure Rates of the hard drives from different brands, then they may checkout the following links -

Failure Rate Analysis - Best 10Tb+ hard drives: Seagate, Western Digital or Toshiba?​


www.backblaze.com

Hard Drive Test Data

Hard Drive test data from the Backblaze data center. Backblaze is affordable, easy-to-use cloud storage.
www.backblaze.com www.backblaze.com


Regards
 
alpha5555 said:
If anyone is concerned about the Reliability Factor and wants to check out the Failure Rates of the hard drives from different brands, then they may checkout the following links -

Failure Rate Analysis - Best 10Tb+ hard drives: Seagate, Western Digital or Toshiba?​

Click to expand...
That often quoted backblaze data is practically irrelevant for a typical consumer. Reason being, unless you are buying hdd with matching model no. down to the last character in that data it is pointless to assume anything based on just partially matching model no. & by the time backblaze publish its data those exact model no. drives can not be purchased typically so only option is to buy used drives matching those models which again makes that data pointless which was for new drives.
 
guest_999 said:
That often quoted backblaze data is practically irrelevant for a typical consumer.
Click to expand...
Yeah, kind of, but still it needs to be often quoted and heeded. With a stance like yours, it is giving rise to confusopoly by manufacturers by releasing a million model numbers and change them as soon as or even before reviews come out. Since the model number proliferation is making other decision making policies difficult, in absence of other criteria, this will have to do. E.g. avoid the closest model number to the bad performing drive from backblaze.
 
kiran6680 said:
Yeah, kind of, but still it needs to be often quoted and heeded. With a stance like yours, it is giving rise to confusopoly by manufacturers by releasing a million model numbers and change them as soon as or even before reviews come out. Since the model number proliferation is making other decision making policies difficult, in absence of other criteria, this will have to do. E.g. avoid the closest model number to the bad performing drive from backblaze.
Click to expand...
That is pure guess work. That's like saying because crucial BX500 is not so good ssd then avoid MX500 too because there is only a difference of one letter. Keep in mind that a typical hdd life depends on many factors outside of manufacturer control which even include which courier company used what route to deliver the hdd to your home. There are some exceptionally & universally bad infamous models in the past but that is just that, in the past & those models were recognized by their name not models anyway (the infamous WD deskstar/deathstar for example or the earliest WD Green models which btw only failed spectacularly in NAS usage leading to starting of nas certified drives trend). Only thing for certain is this, if you don't have backup of your important data in at least 3 different places then no amount of backblaze studies will help you when you really need it.
 
kiran6680 said:
No, educated similarity, of course. Didn't have to be explicitly mentioned in a tech forum, except sometimes, like now.
Click to expand...
Fact is fact, "educated similarity" sounds a lot like "post truth". Whether it is a difference of one letter or more in a predictable pattern or not, the fact is that if model no. is not same down to last character then no objective basis to say they are similar not to mention backblaze run their drives in an environment not found in any buyer's home. Backblaze study is just for reference to see which among seagate/wd/toshiba is currently "trending", that's it.
 
kiran6680 said:
Educated similarity means similarity in one or more factors like manufacturer, controller, nand flash etc.
Click to expand...
You do know that "controller" & "flash" are related to ssd while the discussion is about hdd & all of them are based on same tech (CMR/SMR/HA-SMR) using same material? Anyway I am no longer interested in stretching this off topic discussion so let's end it here.
 
Back
Top