[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: jffs2 flash life calculation



On Fri, 22 August 2003 10:08:56 -0400, Lokesh Kumar wrote:
> 
> Well, thats what I need to determine, but 8 MB is one number.
> 
> On Thu, 2003-08-21 at 15:37, Lokesh Kumar wrote:
> > Roughly 350K of data will get processed every 10 minutes or so, and get
> > removed from the flash. So, there will be multiple files, each of 350K,
> may
> > be 6 of them in the file system at any given time, and then the first 350K
> > file will be processed and removed.
> > 
> > Thanks for your help

Use some assumptions for an estimate:
100,000 erase cycles.
no filesystem overhead.
data doesn't compress at all.
perfect wear levelling.

You write 350kB /10min and need 10 years, 350kB * 60 * 24 * 365 make
something like 183,960MB of data that you write in those ten years.
Divided by the erase cycles it gives you 1.8MB of flash size you
absolutely need.

Less, if the data compresses.
A bit more for file system overhead.
Lots more if you have bad access patterns that create more overhead.
Wear levelling should be near perfect.

So 8MB should be fine, unless the assumptions are too far from
reality.

Jörn

-- 
A victorious army first wins and then seeks battle.
-- Sun Tzu

To unsubscribe from this list: send the line "unsubscribe jffs-dev" in
the body of a message to majordomo@xxxxxxx.com