zen0000 wrote:

And I also read the technical paper about SETI@Home 

"SETI@home: An Experiment in Public-Resource Computing"
in CACM magazine, but in that article they just said 
'Delete work units for which N results has been received'
and 
'Delete work units that have been sent M times'.

What is N and M?

Unknown numerical values.

The boys running SAH probably haven't released the information about how
many times units need to be processed before deletion.