Risks of reformatting applies to SD cards being used on a RPI - thank you for that hint. I was already expecting ext4 to have to do with the problems 
This deals with overclocked RPIs and I read is as this is a special type of corruption only happening with (extremely) overclocked RPIs - I’m not sure whether it’s rather the file system that gets corrupted by overclocked RPIs or the SD cards.
I’m not overclocking my RPI at all.
Thank you very much for the links, especially this one. I was looking for information about how to protect the SD card in a RPI from being used too much and found similar tips.
Unfortunately the best protection seems to be turning off the journaling and I’m not sure I’m going to do that.
I think I try changing the location of the wallet files. Loading them from a tmpfs would be great - for the performance of the flushing process and for the SD card!
edit (edited again, because after restarting the RPI I found out that the folder /tmp/ram got deleted, the entry in /etc/fstab was not working and I was in fact using the manually mounted version of /tmp/ram; creating /ramdisk and entering that in fstab solved that):
I’ve created /tmp/ram /ramdisk, added
tmpfs /ramdisk tmpfs defaults,noatime,nosuid,nodev,noexec,mode=1777,size=16M 0 0
to /etc/fstab, renamed the wallet files, copied them to /ramdisk, created symlinks for walletB.dat and walletS.dat and started nud.
It’s working!
edit2:
Here’s the status of the first flushings after starting nud with the wallets on ram drive; faster than before and not written on SD card!
tail ~/.nu/debug.log -n 500 | grep Flush
DBFlush(false)
DBFlush(true)
2015-09-11 16:35:40 UTC Flushing wallet.dat
Flushed wallet.dat 7ms
2015-09-11 16:36:57 UTC Flushing wallet.dat
Flushed wallet.dat 7ms
2015-09-11 16:37:42 UTC Flushing wallet.dat
Flushed wallet.dat 6ms
