4 ms·
Unexpected Relationship between Hard Drive Life and Temperature
- scott_s 17y agoThe disks that would fail to high temperature early on might have already been weeded out by manufacturer testing. We need to know what sort of quality testing the manufacturers do before making conclusions. Studies on the reliability of computer equipment need to take into account that there is a selection bias going on: equipment on shelves all passed a battery of tests.
- jiaaro 17y agoThat's an interesting thought, but it's still a strange trend
- nandemo 17y agoI'm not sure what you're trying to say. If we wanted to study the physical properties of all hard drives that have been manufactured, then that selection bias might be relevant. However, most of us are only interested in the properties of hardware we can buy.
- frankus 17y agoNote that the graph is average drive temperature. It would be interesting to see at what room temperature drives operating in a server have the lowest failure rate.