On Sunday 28 September 2008 00:18:35 Stroller wrote: > > Since they are not frequently updated and have minimal installed > > software > > (iptables on firewalls and DNS on DNS servers) accompanied by the > > fact that most devices have internal wear leveling; it should take > > many years to reach the write cycle limits? > > I've read a fair little bit about this subject and never gotten a > definitive answer on what is "safe", but AIUI the wear-levelling on > flash memory is filesystem-dependent. Thus it may work fabulously well > for FAT filesystems, and not at all for EXT.
Rule of thumb: The problem is that the ability for individual memory cells to reliably perform writes deteriorates over time. Cheap and nasty devices can start to fail after 10,000 writes to a cell, the better devices can often cope with 100,000 writes to a single cell. The reason there is little definitive data is that it isn't a definitive problem - the variables vary wildly. Like you say, some filesystems do wear levelling (some better than others), some use cases are frugal with their writes, and the device itself has enormous variance as to when it will stop performing as expected. The numbers above must be interpreted as the maximum number of writes where the manufacturer is still prepared to guarantee the device. -- alan dot mckinnon at gmail dot com
