Also bear in mind that since this appears to be “calculating file digests,” this is necessarily an I/O-intensive task, not a CPU-intensive one:   the content of the file must be retrieved from disk in order to be digested.   The ruling constraint will be the capacity of the disk-drive to seek ... to move the read/write head from one track to another.   There will be some opportunity for threading here, but perhaps not nearly so much as you might initially have supposed.