27. omp_dynamic
ā¢Syntax
omp_dynamic = boolean value
ā¢value ā true | false
ā¢True ā allow users to adjust number of threads
ā¢False ā users canāt adjust number of threads
ā¢Default value - false
28. omp_num_threads
ā¢ Syntax
omp_num_threads = num_list
ā¢ Num_list ā positive integer values
ā¢ Single value
ā¢ True & parallel construct without num_threads
ā¢ false & parallel construct without num_threads
ā¢ Multiple values
ā¢ True & parallel construct without num_threads
ā¢ false & parallel construct without num_threads
31. Functions
ā¢ omp_set_num_threads(int num_threads)
ā¢ int omp_get_num_threads
ā¢ int omp_get_max_threads
ā¢ int omp_get_thread_num
ā¢ int omp_get_num_procs
ā¢ void omp_set_dynamic
ā¢ int omp_get_dynamic
ā¢ void omp_set_nested
ā¢ int omp_get_nested
33. Paradigm for using OMP
ā¢ Write a sequential program
ā¢ Identify the portion to be parallelized
ā¢Add directive/pragmas
ā¢ In addition to this call runtime library routines and
modify environment variables
ā¢ Parallel programming is ready.
ā¢ Use OpenMPās compiler to compile
ā¢Run the program
41. Hot research
ā¢Nividia
ā¢Data mining ā tremendous data
ā¢Lacking in techniques and Computational power
ā¢AI/Machine learning
ā¢Image Processing
ā¢Medical Field
ā¢ Image Reconstruction