FilterHN

Flash media longevity testing – 6 years later

67 points

by 1970-01-01

1 day ago

| past

| 7 comments

| old.reddit.com

| HN

▲

ralferoo

58 minutes ago

[-]

I like the fact he's not just verifying all of them each year. AFAICR, reading the flash causes the row to be rewritten with the values just read.

I remember years ago working on the Wii, and there was a restriction on how often you could read the flash to avoid premature wearing. Not sure if that was just the specific type of storage, as googling suggests that NAND is subject to this and NOR isn't. I think pretty much all USB drives now use NOR flash, so maybe this isn't actually an issue any more.

▲

wmf

18 minutes ago

[-]

reading the flash causes the row to be rewritten with the values just read

DRAM works that way but flash doesn't. Read disturb is a different issue.

pretty much all USB drives now use NOR flash

Nope, NOR flash is much more expensive than NAND so NOR is only used for firmware and everything else is NAND.

▲

digdugdirk

42 minutes ago

[-]

What is the best consumer friendly long-term storage medium? Are we still better off with high capacity dvd/Blu ray discs?

▲

1970-01-01

37 minutes ago

[-]

I've been a big fan of M-Disc BD-R.

▲

BoredPositron

29 minutes ago

[-]

What's long-term? I have some dvd-rs that push 20-25 years and despite the plastic getting brittle they still work. I also have some ide drives that still work without problems after 40 years. I would rather aim for 20 years and upgrade the storage device if I still need to retain the data.

▲

vel0city

25 minutes ago

[-]

That's a thought I hadn't had. The plastic of the disk getting so brittle it shatters in the drive due to age. I wonder what's the embrittlement profile of polycarbonate stored in reasonable condition.

▲

ComputerGuru

1 hour ago

[-]

Slightly related: I have a tool that writes random (incompressible) data to a disk and lets you verify it back without storing a copy (by using a csprng seed), initially developed for benchmarking SSDs that used to cheat to get better performance numbers but that can also be used for this purpose or to overwrite (“shred”) a disk: https://github.com/mqudsi/hddrand

▲

fhdkweig

15 minutes ago

[-]

I haven't used badblocks https://en.wikipedia.org/wiki/Badblocks in about 10 years, but I was annoyed that this exact feature wasn't available for testing accidental swapping of block locations. badblocks only writes the same data to each block and thus they are all indistinguishable.

▲

monster_truck

1 hour ago

[-]

Rewriting the data each year hides the actual issue here. Have had plenty of "nice" flash drives rot to hell in 18+ months of dormancy

▲

benterris

1 hour ago

[-]

Does rewriting data help prevent bit rot? Does it mean powered drives can take advantage of it by periodically rewriting the same data over?

▲

monster_truck

16 minutes ago

[-]

It depends on the type of flash being used and the controller managing it. That he did not even identify the chips should inform you of the extent that these results can be trusted.

All I can say for sure is that you should not trust any flash for long term storage, thumb drive or otherwise. In serious enough, high usage, high heat enviornments where everything working without problems or delay is part of what they are paying us to be responsible for, it is standard practice to clone fresh images to nvmes every time, with multiple spares that can be swapped out in minutes when they inevitably fail anyways.

▲

vel0city

21 minutes ago

[-]

It depends on how the flash modules are maintained and their quality, but yes having freshly written data will imply better data consistency on flash media.

Flash media relies on recharging, which may or many not happen often enough.

▲

angry_albatross

1 hour ago

[-]

Did you miss that there are 10 different drives and so they have 10 different years of tests where they are testing a completely untouched drive?

▲

monster_truck

26 minutes ago

[-]

I don't think you're reading the results properly.

▲

thinkling

20 minutes ago

[-]

I think they are reading it correctly. Year 1, they touched one drive and left 9 untouched. Year 2, they read one additional drive and left 8 untouched. Etc.

▲

Springtime

7 minutes ago

[-]

Yes, it's also confirmed on the OP's blog linked in the post.

▲

monster_truck

16 minutes ago

[-]

Those drives aren't being read

▲

nullorempty

1 hour ago

[-]

What's the simplest way to rewrite the data without actually copying the data? Like in place rewrite - you write what you read.

▲

fhdkweig

35 minutes ago

[-]

I've seen "dd if=/dev/removable of=/dev/removable" suggested. I don't know if it actually works or if the OS optimizes it to a no-op.

▲

jmakov

2 hours ago

[-]

Powered all the time on or powered off?

▲

alnwlsn

2 hours ago

[-]

OP says powered off.

▲

01HNNWZ0MV43FF

1 hour ago

[-]

That's good. I want to keep some institutional knowledge and photos in "cold storage" and cloud subscriptions with a credit card and password are completely inviable.

I'll probably get a spinner and a flash drive and hope one of them survives the years.

▲

fhdkweig

29 minutes ago

[-]

If privacy is your primary problem with cloud storage, I would suggest veracrypt containers. And if you aren't storing too much data, I would also suggest DVD/BluRay optical media with DVDisaster and PAR2 archives. I keep a DVD spindle in a safe deposit box that gets updated each year.