[Yaffs] the problem about small data storage

Hi everyone,
I’m doing a research recently about water conservancy informatization.
The monitoring terminal is using ‘linux + yaffs’ system. Since the
rainfall information collected each time is only 20 bytes, yaffs file
system usually applies the ‘out of place’ strategy. I don’t think this
is quite an efficient way. So how about adding ‘partial page
programming’ into yaffs which directly stores new data following the
original without assigning new chunk while small data file is being
written.
I’m wondering if partial page programming can meet the accuracy and
reliability of data storage? Or better solutions to deal with small
data storage would be a great help. I’m open to any kind of
suggestion.
Thanks a lot!