> Do you over-provision your SSD drives to allow for failed block replacement?
> I've owned about 10 Samsung and Intel SSDs over the years and I've yet to have one
> fail. I've had plenty of traditional hard drives shit the bed including a Seagate
> that failed on the first day.
> It's possible that my usage patterns have something to do with my SSD luck. I only
> use the SSD for the OS and programs. I always pair it with a large Western Digital
> green or red 5400 rpm drive for data storage or compiling.
I'd assume the friends buying them follow whatever the recommended practices are, I've tended to avoid them myself due to the horror stories.
I've found with regular HDDs faults either tend to show up early, or they last a long time. Seagates are still by far the worst, the slower speed drives are definitely more reliable.
The main issue with SSDs seems to be that it's always very much 'without warning' Things can seem fine one day, next day you've got a paperweight. No clicking, no sign of failure, no bad SMART reports, just fine one day, dead the next, no chance of recovery.