Parallel similarity computed similarity between given(or all) rows with all other rows of matrix.

Command line options:
--help				- printing help message with list of possible options
--matrix-dir arg		- Path with supermatrix to be computed
--matrix-name arg		- Name of matrix to be computed
--measure-of-relatedness arg	- Measure of relatedness with all parameters for FunctionFactory
--results-file arg		- Ouput file for results
--results-type arg		- Type of storage for results, one of: supermatrix, textfile, densematrix
--read-ccs			- Whether matrix is read from ccs file
--read-crs			- Whether matrix is read from crs file
--minimum-similarity arg	- Minimal value for result to be considered important
--k-best-results arg		- How many results should be stored for each computed row
--target-words-file arg		- File with specific words to be computed

How to run parallel similarity on grid with installed MPI:

mpirun -n arg1 --hostfile arg2 ./parallelSimilarity args

arg1	- Number of nodes, on which we will compute. 1 of nodes is master node dividing work between slaves.
arg2	- File with hostnames of available machines. Should be provided by administrator.
args	- Input arguments for parallelSimilarity

Example how to run:

mpirun -n 24 --hostfile .mpi_hostfile ./parallelSimilarity --matrix-dir /home2/mpiasecki/djaworski/macierze/ --matrix-name kipi_knigi_rzepa_substAdjP_AmIsubj_Coord_NGen_pq --results-type supermatrix --measure-of-relatedness "lin()" --results-file /home2/mpiasecki/djaworski/macierz_podobienstwa_najwieksza_macierz.txt --read-crs --minimum-similarity 0.001 --k-best-results 16000
