Andreas,<br><br>As I understand, the CF cards already implement wear leveling algorithms. Each time a block is updated / written a new block in the CF is used.<br><br>Jordi.<br><br><div class="gmail_quote">2010/7/13 Andreas Delleske <span dir="ltr"><<a href="mailto:delleske@vauban.de">delleske@vauban.de</a>></span><br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">Dear Jordi,<br>
<br>
thank you for your friendly reply!<br>
<div class="im"><br>
> We use adjtimex + ntpd in order to have a ALIX3D2 synchronized. I<br>
> wrote how we configured it in<br>
> <a href="http://code.google.com/p/wfrog/wiki/InstallOn_Alix3D2_voyageLinux0_6_2#Clock_sync" target="_blank">http://code.google.com/p/wfrog/wiki/InstallOn_Alix3D2_voyageLinux0_6_2#Clock_sync</a><br>
<br>
</div>Thanks, I've found that page already.. :-)<br>
<br>
I found out that I could install adjtimex package as soon as I changed<br>
the apt-sources to the german server..<br>
<br>
Now the only buggy package that I need is ssmtp (smtp forwarder to a<br>
smarthost), but i works when ignoring errors when installing.<br>
<div class="im"><br>
> I set up a read-write partition for the database and write a record<br>
> each 5 min. So far the ALIX 3 D 2 with a 1Gb SLC Compact Flash has<br>
> been working for half a year perfectly. We'll have to wait a couple of<br>
> years to know how long the CF last!!!<br>
<br>
</div>Yes..<br>
<br>
If we think about wear leveling, there is a lot of options:<br>
<br>
Here, we will have only tens to hundreds of write updates per day, not<br>
thousands, and I have 4 GByte Flash memory, only 400 MB of which are<br>
used so far. The part where the changes appear (the database) is only<br>
3 MB now and will 10 MB in some years.<br>
<br>
If we assume that every change rewrites 100 kByte (much less in the<br>
database's data, but with every update one or three flash blocks have<br>
to be rewritten..) we rewrite max. 10 MB per day = 3,65 GBytes per<br>
year - within one year, the free space on that disk is rewritten only<br>
once (statistically) that is, if we have a proper wear-leveling<br>
algorithm. If every flash block can be rewritten only 10.000 times,<br>
can we think of 10.000 years? I think that would be enough. :-) Or did<br>
I go wrong?<br>
<br>
Does th CF card "know" which blocks are free and use them<br>
automatically with a round-robin-method? If the CF knows nothing about<br>
filesystems, it would be much less: Assume we write always in the same<br>
10 MB, the CF will last 10.000 days = 27,3 years. Shouldnt even that<br>
be enough?<br>
<br>
I see that if we have one small table in the DB which gets rewritten<br>
always in the same spot, things get worse: 10.000 operations would be<br>
accomplished after only a hundred days... maybe it is crucial to put<br>
only that table in RAM and let the rest of the database grow slowly<br>
without any concern?<br>
<br>
However, it would be great to be able to check the "sanity" of the<br>
CF-disk like with the smartmontools and in some years just swap the<br>
disk (they are inexpensive anyway).<br>
<br>
Or we could move that table from time to time to another location on<br>
the flash ("Manual" wear-leveling)? Lets say we create a copy of the<br>
full database every day and leave to old copy unattended until the<br>
free space is filled up so with every update a new spot is used? Then,<br>
with 3,5 GByte free space divided by 10 MByte a day we'd fill the disk<br>
within a year - but could we redo that 10.000 times? That would be<br>
enough as well.<br>
<br>
--<br>
Kind regards<br>
<font color="#888888">Andreas<br>
</font></blockquote></div><br>