Cyclops Tensor Framework
parallel arithmetic on multidimensional arrays
|
Randomly permuted block write of symmetric matrices from matrix on COMM_SELF to symmetric matrix on COMM_WORLD. More...
Functions | |
int | sparse_permuted_slice (int n, int b, int sym, World &dw) |
tests sparse remote global write via permute function More... | |
char * | getCmdOption (char **begin, char **end, const std::string &option) |
int | main (int argc, char **argv) |
Randomly permuted block write of symmetric matrices from matrix on COMM_SELF to symmetric matrix on COMM_WORLD.
char* getCmdOption | ( | char ** | begin, |
char ** | end, | ||
const std::string & | option | ||
) |
Definition at line 147 of file sparse_permuted_slice.cxx.
Referenced by main().
int main | ( | int | argc, |
char ** | argv | ||
) |
Definition at line 157 of file sparse_permuted_slice.cxx.
References AS, ctf.core::b, getCmdOption(), ctf.core::np(), NS, ctf.core::rank(), SH, sparse_permuted_slice(), and SY.
int sparse_permuted_slice | ( | int | n, |
int | b, | ||
int | sym, | ||
World & | dw | ||
) |
tests sparse remote global write via permute function
[in] | n | dimension of global matrix |
[in] | b | rough dimension of sparse blocks to write from each processor |
[in] | sym | symmetry of the global matrix (and of each block) |
[in] | dw | world/communicator on which to define the global matrix |
Definition at line 18 of file sparse_permuted_slice.cxx.
References ctf.core::b, CTF::Tensor< dtype >::get_local_data(), CTF::Tensor< dtype >::norm2(), ctf.core::np(), NS, CTF::Tensor< dtype >::permute(), ctf.core::rank(), and CTF::Tensor< dtype >::write().
Referenced by main().