Sorry,
/sbin/udevadm info --query=property --name=sdb |grep ID_MODEL
ID_MODEL=M4-CT128M4SSD2
MTBF 1,2 Mio h = 500000 days = 136,98 years
Drive Endurance 72TB=40GB per day for 5 years
I will try to measure the "traffic" next week.
There are Hosted Servers, they are changed every 2 years to "new hardware".
Here are Dedicated Root Servers.
The Webapplication Files (PHP and so on are on RAID), also the Postgres
Data-Dir.
The Root Partition is on SSD, also /boot and swap.
The Images are on SSD, "mounted via bind".
The System has 32GB RAM and thats mainly used for Postgres cache and
MongoDB.
There is a Backend Postgres Database Server that cluster to this
Frontend Server.
So that this Frontend / Application Server can work "standalone".
The mongoDB "cache" the Postgres Querys and some json Objects for the
Application. Serialized Arrays and so on.
When this crash it doesn't matter, It only takes longer to display the
Webpages.
I think 20-30 GB will be free on SSD when it's used as Cache, even more.
(Thats my personal limit, that should be free on system disk.)