I really wish we had better understanding about the failure modes of NVMe past wearing out the flash. Lets say we're writing a vibration sensor's data to the drive, once an hour. We leave 50% of the drive unpartitioned to give the drive controller plenty of headroom for whatever it wants. It might take centuries to get to the expected TBW. Yet there is still some kind of MTBF that is likely to happen.
This above workload scenario is 100% write based. What about drives in data-centers that have lots of user images that they are unlikely/not-incentivized to prune. The drive could, hypothetically, only need to be written once, then forever read. When do sectors start going bad here? Or is the failure mode otherwise, in other componenets? Will this drive last a century? More?
What other modes of failure should we expect, how probable are they?
I love your challenge here @JAlexoid. Switching from thinking in terms of TB/$ to TB/y/$ is a great twist.
This above workload scenario is 100% write based. What about drives in data-centers that have lots of user images that they are unlikely/not-incentivized to prune. The drive could, hypothetically, only need to be written once, then forever read. When do sectors start going bad here? Or is the failure mode otherwise, in other componenets? Will this drive last a century? More?
What other modes of failure should we expect, how probable are they?
I love your challenge here @JAlexoid. Switching from thinking in terms of TB/$ to TB/y/$ is a great twist.