xps150> xdQR_MPI
ScaLAPACK QR factorizations routines.
'for PARKBENCH LINALG kernel suite'
Tests of the parallel real double precision QR factorizations routines.
The following scaled residual checks will be computed:
|| A - QR || / (|| A || * eps * N) and/or
|| A - QL || / (|| A || * eps * N) and/or
|| A - LQ || / (|| A || * eps * N) and/or
|| A - RQ || / (|| A || * eps * N) and/or
|| A - QRP || / (|| A || * eps * N)
The matrix A is randomly generated for each test.
An explanation of the input/output parameters follows:
TIME : Indicates whether WALL or CPU time was used.
M : The number of rows in the matrix A.
N : The number of columns in the matrix A.
MB : The row blocksize of the blocks the matrix A is split into.
NB : The column blocksize of the blocks the matrix A is split into.
P : The number of process rows.
Q : The number of process columns.
THRESH : If a residual value is less than THRESH, CHECK is flagged as PASSED
Fact Time: Time in seconds to factor the matrix.
MFLOPS : Execution rate of the factorization.
The following parameter values will be used:
M : 1000 1000 10000
N : 1000 1000 10000
NB : 6
P : 16
Q : 16
Relative machine precision (eps) is taken to be 0.111022E-15
Routines pass computational tests if scaled residual is less than -3.0000
QR factorization tests.
TIME M N MB NB P Q Fact Time MFLOPS CHECK Residual
---- ------ ------ --- --- ----- ----- --------- ----------- ------ --------
WALL 1000 1000 6 6 16 16 6.36 210.12 BYPASS 0.0E+00
WALL 1000 1000 6 6 16 16 3.25 410.94 BYPASS 0.0E+00
WALL 10000 10000 6 6 16 16 166.20 8023.73 BYPASS 0.0E+00
Finished 3 tests, with the following results:
3 tests completed without checking.
0 tests skipped because of illegal input values.
END OF TESTS.
FORTRAN STOP