Excellent high capacity drives, especially when torn from the enclosure and used in a NAS (without RAID).
25 comments
simonprr
15 Aug 17#1
Nice price... why do u say without RAID?
cd0nc to simonprr
15 Aug 17#3
The drives use Shingled Magnetic Recording (SMR) tech and are geared towards sequential writes
FireOnAWire to cd0nc
15 Aug 17#13
Are you sure it's SMR? I have a couple of these and the write speed doesn't drop after a while like other SMR drives. Even when I was transferring 300GB in one sitting. Perhaps they've upped the buffer or refined the process so they're less of a poor performer.
cd0nc to FireOnAWire
15 Aug 17#6
I've got about 6 of these already and I hear what you're saying. The launch units seem to exhibit a noticeable drop after the cache filled up. The two newer units when I was filling them achieved a much higher sustained write speed and for longer. I changed my storage controller so chalked it up to that.
I've not done anything RAID related at home for many years, YMMV, worth reading the comments of people using RAID 6, ZFS etc. with no issues over at Reddit perhaps. Personally I run my RAID controller in IT mode and everything as a single disks, mapping disks to paths. I watch out for tell-tale signs such as SMART errors, long pauses and replace drives as they get issues or I buy bigger ones. If something does go wrong I've found it's much more straightforward to run recovery software on a single disk. I don't need the redundancy and anything important is properly backed up.
CampGareth to cd0nc
15 Aug 17#8
Ceph is something you should look into as each drive is in JBOD mode with software above that talking to other drives, handling replication etc. meaning if a drive should fail the others can figure out what data's not got the desired number of copies and will start copying it to more. Better yet it understands failure domains so if you have two machines it'll try to make sure that copies aren't stored alongside the original on the same machine. I'm using it at home as a single-box NAS until I have money for more machines and drives but it handled a recent drive failure without me even noticing.
FireOnAWire to cd0nc
15 Aug 17#14
Yeah I had one of the early Archive branded 8TB drives and after copying over around 60GB of data the write speed would tank to 40MB/s. Teracopy was able to get around the limitation somehow but I wasn't a fan of the application.
With these drives I am getting 150-170MB/s sustained writes. These drives seem to behave like regular hard drives. Well for data storage purposes anyway, I've not tried using them for anything else.
popolou to cd0nc
16 Aug 17#17
What could be more straight forward than replacing a failed drive and allowing the raid system rebuild it back into the array? :astonished:
Pops
fat2fit to popolou
16 Aug 17#18
If it is replaceable data such as media/ISOs then there is nothing wrong with avoiding RAID. You might have to replace the data on one disk but you avoid the issues of losing an entire array if a drive fails while you rebuild your array. Using something like stablebit to scan your drives automatically for errors should help avoid dataloss too
popolou to fat2fit
16 Aug 17#21
There's a bit of flip flopping in the logic there.
Isn't the point really that all the faffing about needed so as to purposely avoid raid can be massively offset with the ease of recovery and benefits of a suitable raid setup?
I can lose two drives in one of my arrays (that took all of only a few minutes to create) before the container is degraded and not have to faff around with any of the alternatives suggested here.
Pops
uni to cd0nc
16 Aug 17#20
this isn't necesarily true. i got one in the £120 amazon deal and it was a PMR drive. i was expecting SMR, but after a bit of googling i found a US based forum where i found some people got SMR drives and some people got PMR drives, but until you get it you can't tell what you are going to get
it's possible they started using SMR drives and changed to PMR or vice versa or just stuck whatever they got in depending on batches of manufacture. but i was pleasantly surprised to find a 7200rpm PMR drive that filled in a decent time
cd0nc to uni
16 Aug 17#22
Looks like after checking, both of the drives I got last time are PMR. Not that I've got anything against the old one, but like your say, it's nice to have these filling up in a decent amount of time.
I don't have any need for redundancy. You're alluding I should use it as a backup solution, which from my experience is prone to failure and not what RAID is designed for. I've watched two disks bought at a similar time take down an array for good, at which point I had to restore from tape.
popolou to cd0nc
16 Aug 17#23
Well, no and I didn't. You suggested that "if something does go wrong I've found it's much more straightforward to run recovery software on a single disk" for which I took issue with. You're pitting your odds of recovering a failed disk and the very considerable time it takes to do that post-incident (not least the slim success of it happening or assuming it is a recoverable failure at all) against the simple task of configuring an array upfront for redundancy and prevent any and all of that risk. Of course, should you have very specific requirements i suspect you have good reason why you rely on a JBOD but i am presuming this is not the case from what i've read. So, anyone faced with identical options would have a very clear decision to make. Frankly, your reactive recovery scenario is frankly unworkable.
You previously alluded to the Raid 5 black hole and now the story of rebuilding two disks before seeing the array fail. None of these incidents would happen with a properly & suitably configured array or with the right knowledge.
Pops
benjai
15 Aug 17#2
Coming down gradually now although not as good as the £120 a while back. Are prices for large HDDs finally coming down in price?
CyDoNiA
15 Aug 17#4
Awaiting the flood of comments on how bad Seagate drives are... :wink:
jamgin to CyDoNiA
15 Aug 17#10
Seagate are lovely
4Real2016 to CyDoNiA
15 Aug 17#11
Seagate high capacity drives have a failure rate that's on par with WD.
It's a shame WD bought HGST as there drives were incredibly reliable.
I've bought this drive before and it had a SMART error after 30mins of
use. It failed with a temperature warning - which seems to be common
error with most Seagate drives. I ended up returning it and went with WD
instead.
It's not worth risking 8TB of data on a Seagate IMO.
cpioi
15 Aug 17#12
Sorry to mention this but bought this on prime day for 119.99 about 5 to 6 weeks ago. Fast, reliable, would buy again.
LocoMoFo9999 to cpioi
16 Aug 17#15
Wait a bit longer :relaxed:
pantaiema
16 Aug 17#16
Shown 179.99. How did you get it £165.98
pantaiema
16 Aug 17#19
179.99 now ?? How to get it down to £165.98 any vode ??
zebrum
3 Sep 17#24
Seems the new models of these use ST8000DM004 which although very similar model number is less reliable, slightly slower but uses less power than the ST8000DM005 Barracuda Pro and yes these don't use shingled recording (SMR).
bens123
4 Sep 17#25
I bought a load of Seagates ranging from 2TB to 3TB and every single one failed prematurely. Since then I have always bought WD and they have been going strong for years.
![Fresh Breakfast Meat Box (Using code) @ Campbells Meat [Haggis / Lorne Sausage / Bacon / Black Pudding & More]](/images/topics/208684_1.jpg)
![[Xbox One] Q. u. b. e: Director's Cut on Deals with Gold](/images/topics/208716_1.jpg)
Opening post
25 comments
I have a couple of these and the write speed doesn't drop after a while like other SMR drives. Even when I was transferring 300GB in one sitting.
Perhaps they've upped the buffer or refined the process so they're less of a poor performer.
I've not done anything RAID related at home for many years, YMMV, worth reading the comments of people using RAID 6, ZFS etc. with no issues over at Reddit perhaps.
Personally I run my RAID controller in IT mode and everything as a single disks, mapping disks to paths. I watch out for tell-tale signs such as SMART errors, long pauses and replace drives as they get issues or I buy bigger ones. If something does go wrong I've found it's much more straightforward to run recovery software on a single disk. I don't need the redundancy and anything important is properly backed up.
With these drives I am getting 150-170MB/s sustained writes.
These drives seem to behave like regular hard drives. Well for data storage purposes anyway, I've not tried using them for anything else.
Pops
Isn't the point really that all the faffing about needed so as to purposely avoid raid can be massively offset with the ease of recovery and benefits of a suitable raid setup?
I can lose two drives in one of my arrays (that took all of only a few minutes to create) before the container is degraded and not have to faff around with any of the alternatives suggested here.
Pops
it's possible they started using SMR drives and changed to PMR or vice versa or just stuck whatever they got in depending on batches of manufacture. but i was pleasantly surprised to find a 7200rpm PMR drive that filled in a decent time
I don't have any need for redundancy. You're alluding I should use it as a backup solution, which from my experience is prone to failure and not what RAID is designed for. I've watched two disks bought at a similar time take down an array for good, at which point I had to restore from tape.
You previously alluded to the Raid 5 black hole and now the story of rebuilding two disks before seeing the array fail. None of these incidents would happen with a properly & suitably configured array or with the right knowledge.
Pops
It's a shame WD bought HGST as there drives were incredibly reliable.
It's not worth risking 8TB of data on a Seagate IMO.