| 1 |
On Sun, 30 May 2010 11:48:21 +0100 |
| 2 |
Neil Bothwick wrote: |
| 3 |
|
| 4 |
> On Sat, 29 May 2010 07:59:31 -0400, David Relson wrote: |
| 5 |
> |
| 6 |
> > Indeed flash drives _do_ have a lifetime. My recollection is that |
| 7 |
> > it's in the thousands of writes if not the hundreds of thousands of |
| 8 |
> > writes. Assuming a life of 1,000 writes and you backup once daily, |
| 9 |
> > that's 3 years of backups. 10,000 writes would be 30 years. Of |
| 10 |
> > course if you backup every hour, 10,000 writes is a year (or so). |
| 11 |
> |
| 12 |
> You're assuming that each backup only writes once, which is far from |
| 13 |
> true. If you mount a drive with the sync option, the FAT is updated |
| 14 |
> for every block you write, so even a single file can cause thousands |
| 15 |
> of writes to the same location. |
| 16 |
> |
| 17 |
> |
| 18 |
> -- |
| 19 |
> Neil Bothwick |
| 20 |
|
| 21 |
Neil, |
| 22 |
|
| 23 |
Correct -- if the USB is mounted synchronously. |
| 24 |
|
| 25 |
Normally Linux uses asynchronous writes (caching), so will hit the FAT |
| 26 |
much less often. |
| 27 |
|
| 28 |
I've tried synchronous writes and it's a real performance killer. |
| 29 |
|
| 30 |
However for a DOS formatted stick (which is the norm) the FAT does seem |
| 31 |
to be the week link. |
| 32 |
|
| 33 |
David |
Archives