Wondering about load...
- Thread starter TErxleben
- Start date
It always depends on what you are doing and how much you care about your data and downtimes. If you think the performance and interpolated life expectation is fine for your usecase, then it's all good.
For me those wouldn't be able to handle the workload of my homelab (but I also got ~40 guests and DBs that are writing ~1TB to the SSDs each day). I personally replaced all the EVO M.2 SSDs, other consumer SSDs and HDDs with proper MLC enterprise SSDs, because they were always failing or couldn't handle the sync writes.
With SSDs, most of the time, you get what you pay for. If you for example pay 10 times the price but then its 100x as faster and will last 30x longer, thats still a great deal.
For me those wouldn't be able to handle the workload of my homelab (but I also got ~40 guests and DBs that are writing ~1TB to the SSDs each day). I personally replaced all the EVO M.2 SSDs, other consumer SSDs and HDDs with proper MLC enterprise SSDs, because they were always failing or couldn't handle the sync writes.
With SSDs, most of the time, you get what you pay for. If you for example pay 10 times the price but then its 100x as faster and will last 30x longer, thats still a great deal.
Last edited:
running regular ssd on ext4 is reliable because wear level isn't over burned.
make sure backups are done because in rare case, eg. power failure or system crash, data can be loss because drive internal "cache" can't be flushed because of lack of power protection like regular desktop hw / workstation hw. filesystem isn't broken because a journal help to know where to restart.
imo, in PVE, where filesystems are nested, it's a bit amplified. VM Cache vDisk is set to None to be more safe, but lack some top speeds.
(sorry for my english..)
make sure backups are done because in rare case, eg. power failure or system crash, data can be loss because drive internal "cache" can't be flushed because of lack of power protection like regular desktop hw / workstation hw. filesystem isn't broken because a journal help to know where to restart.
imo, in PVE, where filesystems are nested, it's a bit amplified. VM Cache vDisk is set to None to be more safe, but lack some top speeds.
(sorry for my english..)