Disk file systems can become very fragmented. Large files such as
history and restart files are slow to read or write when this
happens. Consequently, for large datasets the model
pre-allocates space equal to the expected
length of the output file using the setf command with the -c
option. Routine prealc invokes setf to find as
much contiguous space as possible for large output file.
Questions on these pages can be sent to... erik@ucar.edu .
Search page