7 ms·

I built a 2 drive 4TB RAID1 with a DS220j using these SMR drives the other day. To be honest performance is as I'd expect out of a hard drive. But I've mostly

by Jonnax 6y ago

I built a 2 drive 4TB RAID1 with a DS220j using these SMR drives the other day. To be honest performance is as I'd expect out of a hard drive. But I've mostly been using SSDs on my PCs. I've put my instrument VSTs on it and I'm using the Windows NFS client (you use cli command to mount after enabling in add windows features) Standard windows file sharing (SMB) seems to top out at 25MB/s and I've no idea why. I'm not too worried about it being SMR at least looking at my use case. But it's really messed up that WD hid the technology used. If anything it would drive sales for their more expensive products. Also it does highlight that in a lot of tech product fields we don't have good reviews. Because there's no money in it. Like with laptops there's notebookcheck doing technical reviews. With displays rtings and tftcentral. But with other less used stuff it's a YouTube video from someone saying "yeah it's a harddrive" and reading the marketing. Or a drive will get reviewed. Then the manufacture will start selling something with the same brand but different model number to confuse customers. But nobody will re-review, partially because they weren't told.

dannyw 6y ago
Heh, that's almost the same as my setup, except I bought the Seagate Ironwolfs instead on sale. I paid AUD$20 more in total for the Ironwolfs. I am looking to sell my DS220j however. The lack of support for btrfs is very unfortunate, as a NAS shouldn't be using ext4 with zero integrity measures and zero bit-rot protection in 2020. RAID1 will not save anything as if there is bitrot, without checksums there is no way to know which copy is correct. I'm looking to replace the DS220j with the DS220+ when it comes out, which is supposedly in just 1 week based on a local PC shop I called. I also might upgrade to the DS420+ depending on pricing.
- ksec 6y ago
  >The lack of support for btrfs is very unfortunate, as a NAS shouldn't be using ext4 with zero integrity measures and zero bit-rot protection in 2020. Yes. But 220+ is way more expensive than 220J. For Something I expect in 2020 should be standard feature. And those Ironwolf HDD aren't cheap either. The way things are going Cloud Storage is an increasingly attractive option.
  - dannyw 6y ago
    Yeah, I agree. It sucks how the 220j doesn't have such a basic feature. Although, I was comparing costs with a self built Pi 4 with 2x HDD enclosure and powered USB hub, and the 220j stacks up favorably on the pricing front. You also get the plug in play setup, DSM software, and support. The DS220+ does not stack up favorably on the pricing front though.
ubercow13 6y ago
If you just built your array, you wouldn't encounter this issue yet. The problem would come (assuming you're using ZFS) when your NAS is full of data and one of your disks fails, and you try to replace it with another SMR drive. The drive would be rejected before it could be added to the array.
- dannyw 6y ago
  As benchmarked in the article, SMR drives rebuild Linux RAID fine because it's just a sequential data overwrite. The problem occurs when you are dealing with sequential files with small sector sizes, and non-sequential files, where performance is abysmal: 93% slower than CMR as benchmarked by servethehome and Ars.
  - zozbot234 6y ago
    The problem is not just the slow performance, but the fact that these drives can be dropped from the array when rebuilding - RAID implementations including the one in ZFS treat abysmal performance as damage and route around it. The performance problem could be addressed simply by raising the number of drives in the array (RAID5 to RAID6 or more).
    - ezoe 6y ago
      Isn't that a problem of ZFS? I don't mind SMR HDD if it was labeled as such and warn about it's weird performance traits.
      - ubercow13 6y ago
        This whole issue was originally that Western Digital was trying to hide which HDDs were SMR. ZFS can't be changed to have special behaviour for SMR drives if the vendors try to hide the fact they are SMR, even from software. It wouldn't be able to identify them.
        ezoe 6y ago

by Jonnax 6y ago

dannyw 6y ago

Heh, that's almost the same as my setup, except I bought the Seagate Ironwolfs instead on sale. I paid AUD$20 more in total for the Ironwolfs. I am looking to sell my DS220j however. The lack of support for btrfs is very unfortunate, as a NAS shouldn't be using ext4 with zero integrity measures and zero bit-rot protection in 2020. RAID1 will not save anything as if there is bitrot, without checksums there is no way to know which copy is correct. I'm looking to replace the DS220j with the DS220+ when it comes out, which is supposedly in just 1 week based on a local PC shop I called. I also might upgrade to the DS420+ depending on pricing.

ksec 6y ago
>The lack of support for btrfs is very unfortunate, as a NAS shouldn't be using ext4 with zero integrity measures and zero bit-rot protection in 2020. Yes. But 220+ is way more expensive than 220J. For Something I expect in 2020 should be standard feature. And those Ironwolf HDD aren't cheap either. The way things are going Cloud Storage is an increasingly attractive option.
- dannyw 6y ago
  Yeah, I agree. It sucks how the 220j doesn't have such a basic feature. Although, I was comparing costs with a self built Pi 4 with 2x HDD enclosure and powered USB hub, and the 220j stacks up favorably on the pricing front. You also get the plug in play setup, DSM software, and support. The DS220+ does not stack up favorably on the pricing front though.

ubercow13 6y ago

If you just built your array, you wouldn't encounter this issue yet. The problem would come (assuming you're using ZFS) when your NAS is full of data and one of your disks fails, and you try to replace it with another SMR drive. The drive would be rejected before it could be added to the array.

dannyw 6y ago
As benchmarked in the article, SMR drives rebuild Linux RAID fine because it's just a sequential data overwrite. The problem occurs when you are dealing with sequential files with small sector sizes, and non-sequential files, where performance is abysmal: 93% slower than CMR as benchmarked by servethehome and Ars.
- zozbot234 6y ago
  The problem is not just the slow performance, but the fact that these drives can be dropped from the array when rebuilding - RAID implementations including the one in ZFS treat abysmal performance as damage and route around it. The performance problem could be addressed simply by raising the number of drives in the array (RAID5 to RAID6 or more).
  - ezoe 6y ago
    Isn't that a problem of ZFS? I don't mind SMR HDD if it was labeled as such and warn about it's weird performance traits.
    - ubercow13 6y ago
      This whole issue was originally that Western Digital was trying to hide which HDDs were SMR. ZFS can't be changed to have special behaviour for SMR drives if the vendors try to hide the fact they are SMR, even from software. It wouldn't be able to identify them.
      - ezoe 6y ago