It should be possible to get an aggregated total value across dynamic selections from the OLAP server.
Example: get a total revenue value accross selected products matching a certain criteria.
So it should be possible to get an aggregated total of a subset out of a single PALO.DATA formula,
like this:
PALO.DATA ("myDB","ProductCube",PALO.SUBSET(), ...)
instead of only getting the value for dimension elements via:
PALO.DATA ("myDB","ProductCube","ProductGroupA", ...)

I finally found a way to do that with 2 subsets :

I have 9 dimensions in my cube and I wanted to use 2 subsets in 1 formula. By playing with curly brackets I obtain the good result by using the formula below :

=SUM(PALO.DATAV("ServerName","CubeName",

"ValueDimension1","ValueDimension2",

{PALO.SUBSET("ServerName","ValueDimension3",1,,PALO.HFILTER(,0,FALSE,,,,,2,2,FALSE),PALO.TFILTER("^A|^B|^C|^D|^E",1,FALSE),,,,PALO.SORT(1,0,,0,,0,1,,))},"ValueDimension4","ValueDimension5","ValueDimension6","ValueDimension7","ValueDimension8",

PALO.SUBSET("ServerName","ValueDimension9",1,,PALO.HFILTER(,0,FALSE,2,,,,,,),PALO.TFILTER({"^A|^B|^C|^D|^E|^F|^G|^H|^I|^J|^K|^L|^M|^N|^O|^P"},1,FALSE),,,,PALO.SORT(1,0,,0,,0,1,,))))Next step with 3 subsets !! ^^

It's possible to do that by combinate SUM(PALO.DATAV("myDB","ProductCube",PALO.SUBSET(),...)) functions but its seems to be limited using only one subset