5 ms·

Is it going to be hamstrung by SATA III? It would take roughly 13 hours to fill the drive at 6 Gbps…

wtallis 3y ago
Only if it's a multi-actuator drive. Single-actuator drives are still barely hitting SATA II speeds, so this density boost on its own probably won't be enough to increase performance to the point where SATA III's 6Gbps is a real problem.
dsr_ 3y ago
You get some data throughput increases by increasing areal density and number of platters -- the days of counting on 100-120MB/s from a spinner are now replaced by checking to see whether you get 220-280MB/s -- but you should not expect the rotating disk technology to push the SATA III standard just yet. It will be more than 13 hours.
lostlogin 3y ago
Adding an 18TB to a small Synology array takes days as it is. Adding a drive this size would take at least a week I think.
- Hamuko 3y ago
  The biggest problem that I have with adding big drives to my NAS is the amount of prep work. badblocks takes fucking forever against bigger drives. It was like, 5 days to run three passes of badblocks against a 12 TB drive or something?
  - bombcar 3y ago
    Is it even worth running something like badblocks before adding drives anymore? I guess if you want to be absolutely sure it's "clean" but most drives do intelligent remapping internally anyway, so "add and scrub afterwards" may be the way to go.
    - Hamuko 3y ago
      I feel like it's less of a bother to diagnose if the drive needs to be returned immediately and not after you've already slapped it into your RAID.
gnicholas 3y ago
Ah yes, "mo TB, mo problems", as the Notorious B.I.G would lament.
codemac 3y ago
You are hitting on the largest problem with high density HDDs - the IO dosn't scale with the density, and thus the data needs to be colder and colder that you put on the drive. The cold data thesis hasn't worked out as well in the market as many storage vendors would have liked.
- nayuki 3y ago
  There is a physical reason why I/O speed doesn't scale with density. When the density of data on disks gets denser, it does so on both axes - x and y, or radial and tangential. But read/write heads only access one track at a time, so the speed-up is only along one axis. Think about CDs, DVDs, and BDs. The discs can only be spun at a finite rate (somewhere around 10k RPM) before they shatter. The R/W head can read the data within any single track at full speed, but only one track at a time. From CDs to BDs, the amount of data per track increased (this doesn't affect the amount of time needed to read all the data from a disc), but also the number of tracks increased (this does affect the total time).

Only if it's a multi-actuator drive. Single-actuator drives are still barely hitting SATA II speeds, so this density boost on its own probably won't be enough to increase performance to the point where SATA III's 6Gbps is a real problem.

dsr_ 3y ago

You get some data throughput increases by increasing areal density and number of platters -- the days of counting on 100-120MB/s from a spinner are now replaced by checking to see whether you get 220-280MB/s -- but you should not expect the rotating disk technology to push the SATA III standard just yet. It will be more than 13 hours.

lostlogin 3y ago

Adding an 18TB to a small Synology array takes days as it is. Adding a drive this size would take at least a week I think.

Hamuko 3y ago
The biggest problem that I have with adding big drives to my NAS is the amount of prep work. badblocks takes fucking forever against bigger drives. It was like, 5 days to run three passes of badblocks against a 12 TB drive or something?
- bombcar 3y ago
  Is it even worth running something like badblocks before adding drives anymore? I guess if you want to be absolutely sure it's "clean" but most drives do intelligent remapping internally anyway, so "add and scrub afterwards" may be the way to go.
  - Hamuko 3y ago
    I feel like it's less of a bother to diagnose if the drive needs to be returned immediately and not after you've already slapped it into your RAID.

gnicholas 3y ago

Ah yes, "mo TB, mo problems", as the Notorious B.I.G would lament.

codemac 3y ago

You are hitting on the largest problem with high density HDDs - the IO dosn't scale with the density, and thus the data needs to be colder and colder that you put on the drive. The cold data thesis hasn't worked out as well in the market as many storage vendors would have liked.

nayuki 3y ago
There is a physical reason why I/O speed doesn't scale with density. When the density of data on disks gets denser, it does so on both axes - x and y, or radial and tangential. But read/write heads only access one track at a time, so the speed-up is only along one axis. Think about CDs, DVDs, and BDs. The discs can only be spun at a finite rate (somewhere around 10k RPM) before they shatter. The R/W head can read the data within any single track at full speed, but only one track at a time. From CDs to BDs, the amount of data per track increased (this doesn't affect the amount of time needed to read all the data from a disc), but also the number of tracks increased (this does affect the total time).