| Back to the main page of this manual | Input reference of CP2K version 2.3 (Revision 12341) |
| MEMORY 0 1 2 NMPI | |
| The memory banks ID for each MPI process. It will beuse to set memory bank of the MPI process | |
| This optional keyword cannot be repeated and it expects a list of integers. | |
| Default value: 0 |
| MEM_POL TYPE (D=DEFAULT, L=LOCAL, I=INTERLEAVE, M=MANUAL ) | |
| Enable memory binding on the compute node. | |
| This optional keyword cannot be repeated and it expects precisely one word. | |
| Default value: X |
| MPI_REORDERING TYPE (D=NONE H=HILBERT P=PEANO N=SNAKE CURVE R=ROUND-ROBIN, F=HILBERT-PEANO, S=SWITCH PAIRS, C=CANNON, O=OWN ORDERING) | |
| Apply a global MPI reordering for the run | |
| This optional keyword cannot be repeated and it expects a list of words. | |
| Default value: N |
| PRINT_BRANCH TRUE | |
| Print machine architecture component organization. | |
| This optional keyword cannot be repeated and it expects precisely one logical. | |
| Default value: .FALSE. |
| PRINT_FULL TRUE | |
| Print full machine architecture | |
| This optional keyword cannot be repeated and it expects precisely one logical. | |
| Default value: .FALSE. |
| PRINT_PROC TRUE | |
| Print the process binding in the machine. | |
| This optional keyword cannot be repeated and it expects precisely one logical. | |
| Default value: .FALSE. |
| PRINT_RESUME TRUE | |
| Print the compute node architecture number of components. | |
| This optional keyword cannot be repeated and it expects precisely one logical. | |
| Default value: .FALSE. |
| PRINT_THREAD TRUE | |
| Print the threads binding in the machine. | |
| This optional keyword cannot be repeated and it expects precisely one logical. | |
| Default value: .FALSE. |
| PRINT_THREAD_CUR TRUE | |
| Print the current threads binding in the machine. | |
| This optional keyword cannot be repeated and it expects precisely one logical. | |
| Default value: .FALSE. |
| PROCESS 0 1 2 NMPI | |
| Core ID for each MPI process of a compute node. It will be used to place the MPI process. | |
| This optional keyword cannot be repeated and it expects a list of integers. | |
| Default value: 0 |
| SCHED_MPI TYPE (N=NONE L=LINEAR I=INTERLEAVED, G=GROUP, M=MANUAL) | |
| Enable process scheduling on the compute node. | |
| This optional keyword cannot be repeated and it expects precisely one word. | |
| Default value: D |
| SCHED_THREAD TYPE (N=NONE L=LINEAR I=INTERLEAVED, G=GROUP,M=MANUAL) | |
| Enable thread scheduling on the compute node. | |
| This optional keyword cannot be repeated and it expects precisely one word. | |
| Default value: D |
| THREAD 0 1 2 NTHREADS | |
| Core ID for each thread of a compute node. It will be used to place the thread. | |
| This optional keyword cannot be repeated and it expects a list of integers. | |
| Default value: 0 |
| Back to the main page of this manual or the CP2K home page | (Last update: 3.9.2012) |