Imagine that your tons of terrabytes of data can be processed in space or time
(or combination) with only a few readable(!) lines of code. On the one hand, we
would obviously fall back to tools like cdo or ncl for such tasks. However, we
are stuck to the limited data analysis options that these provide. On the other
hand, scientific software packages like Matlab, R and IDL (and also Python of
course!) provide very customizable climate data analysis tools, yet they have
severe memory restrictions due to the limited RAM in computers. They also don't
allow a quick moviewize visualization of such large data sets, simply because
they cannot load all those 'terrabytes' into memory at once. The
PynaColaDa-tool now provides the best of both worlds: it allows to perform any
arbitrary pre-defined or user-defined custom function analysis on a massively
HUGE dataset very easily, with great performance (from the numpy/matlib
library), and WITHOUT the memory restrictions! As it both directly 'reads from'
and 'writes to' NetCDF files, it allows to easily scroll through your analysis
data sets, i.e. with ncview.