|
3434 | 3434 | <BATCH_SYSTEM>pbspro</BATCH_SYSTEM> |
3435 | 3435 | <SUPPORTED_BY>e3sm</SUPPORTED_BY> |
3436 | 3436 | <MAX_TASKS_PER_NODE>208</MAX_TASKS_PER_NODE> |
3437 | | - <MAX_TASKS_PER_NODE compiler="oneapi-ifxgpu">96</MAX_TASKS_PER_NODE> |
| 3437 | + <MAX_TASKS_PER_NODE compiler="oneapi-ifxgpu">12</MAX_TASKS_PER_NODE> |
3438 | 3438 | <MAX_MPITASKS_PER_NODE>104</MAX_MPITASKS_PER_NODE> |
3439 | | - <MAX_MPITASKS_PER_NODE compiler="oneapi-ifxgpu">48</MAX_MPITASKS_PER_NODE> |
| 3439 | + <MAX_MPITASKS_PER_NODE compiler="oneapi-ifxgpu">12</MAX_MPITASKS_PER_NODE> |
3440 | 3440 | <PROJECT_REQUIRED>FALSE</PROJECT_REQUIRED> |
3441 | 3441 | <mpirun mpilib="default"> |
3442 | 3442 | <executable>mpiexec</executable> |
3443 | 3443 | <!--executable>numactl -m 2-3 mpiexec</executable--><!--for HBM runs--> |
3444 | 3444 | <arguments> |
3445 | 3445 | <arg name="total_num_tasks">-np {{ total_tasks }} --label</arg> |
3446 | 3446 | <arg name="ranks_per_node">-ppn {{ tasks_per_node }}</arg> |
3447 | | - <arg name="ranks_bind">-envall</arg> |
| 3447 | + <arg name="ranks_bind">--cpu-bind $ENV{RANKS_BIND}</arg> |
3448 | 3448 | <arg name="threads_per_rank">-d $ENV{OMP_NUM_THREADS}</arg> |
3449 | 3449 | <arg name="gpu_maps">$ENV{GPU_TILE_COMPACT}</arg> |
3450 | 3450 | </arguments> |
|
3462 | 3462 | <modules compiler="!gnu"> |
3463 | 3463 | <command name="load">oneapi/eng-compiler/2024.07.30.002</command> |
3464 | 3464 | </modules> |
3465 | | - <modules compiler="oneapi-ifxgpu"> |
3466 | | - <command name="load">kokkos/4.4.01-omp-sycl</command> |
3467 | | - </modules> |
| 3465 | + <!--<modules compiler="oneapi-ifxgpu"> --> |
| 3466 | + <!-- <command name="load">kokkos/4.4.01-omp-sycl</command> --> |
| 3467 | + <!--</modules> --> |
3468 | 3468 | <modules compiler="gnu"> |
3469 | 3469 | <command name="unload">spack-pe-gcc cmake</command> |
3470 | 3470 | <command name="load">gcc/10.3.0</command> |
|
3478 | 3478 | <env name="PNETCDF_PATH">/lus/flare/projects/E3SM_Dec/soft/pnetcdf/1.14.0/oneapi.eng.2024.07.30.002</env> |
3479 | 3479 | <env name="LD_LIBRARY_PATH">/lus/flare/projects/E3SM_Dec/soft/pnetcdf/1.14.0/oneapi.eng.2024.07.30.002/lib:/lus/flare/projects/E3SM_Dec/soft/netcdf/4.9.2c-4.6.1f/oneapi.eng.2024.07.30.002/lib:$ENV{LD_LIBRARY_PATH}</env> |
3480 | 3480 | <env name="PATH">/lus/flare/projects/E3SM_Dec/soft/pnetcdf/1.14.0/oneapi.eng.2024.07.30.002/bin:/lus/flare/projects/E3SM_Dec/soft/netcdf/4.9.2c-4.6.1f/oneapi.eng.2024.07.30.002/bin:$ENV{PATH}</env> |
| 3481 | + <env name="RANKS_BIND">list:0-7,104-111:8-15,112-119:16-23,120-127:24-31,128-135:32-39,136-143:40-47,144-151:52-59,156-163:60-67,164-171:68-75,172-179:76-83,180-187:84-91,188-195:92-99,196-203</env> |
3481 | 3482 | </environment_variables> |
3482 | 3483 | <environment_variables DEBUG="TRUE"> |
3483 | 3484 | <env name="HYDRA_TOPO_DEBUG">1</env> |
3484 | 3485 | </environment_variables> |
3485 | 3486 | <environment_variables compiler="oneapi-ifxgpu"> |
3486 | 3487 | <env name="ONEAPI_DEVICE_SELECTOR">level_zero:gpu</env> |
3487 | | - <env name="ONEAPI_MPICH_GPU">NO_GPU</env> |
3488 | | - <env name="MPIR_CVAR_ENABLE_GPU">0</env> |
| 3488 | + |
| 3489 | + <!-- <env name="ONEAPI_MPICH_GPU">NO_GPU</env> --> |
| 3490 | + |
| 3491 | + <env name="MPIR_CVAR_CH4_COLL_SELECTION_TUNING_JSON_FILE"></env> |
| 3492 | + <env name="MPIR_CVAR_COLL_SELECTION_TUNING_JSON_FILE"></env> |
| 3493 | + <env name="MPIR_CVAR_CH4_POSIX_COLL_SELECTION_TUNING_JSON_FILE"></env> |
| 3494 | + <env name="UR_L0_USE_DRIVER_INORDER_LISTS">1</env> |
| 3495 | + <env name="UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS">1</env> |
| 3496 | + <env name="UR_L0_USE_COPY_ENGINE_FOR_IN_ORDER_QUEUE">1</env> |
| 3497 | + <!--<env name="FI_PROVIDER">cxi</env>--> |
| 3498 | + <env name="FI_MR_CACHE_MONITOR">disabled</env> |
| 3499 | + <env name="FI_CXI_OVFLOW_BUF_SIZE">8388608</env> |
| 3500 | + <env name="PALS_PING_PERIOD">240</env> |
| 3501 | + <env name="PALS_RPC_TIMEOUT">240</env> |
| 3502 | + |
| 3503 | + <env name="MPIR_CVAR_ENABLE_GPU">0</env> |
3489 | 3504 | <env name="romio_cb_read">disable</env> |
3490 | 3505 | <env name="romio_cb_write">disable</env> |
3491 | 3506 | <env name="SYCL_CACHE_PERSISTENT">1</env> |
3492 | 3507 | <env name="GATOR_INITIAL_MB">4000MB</env> |
3493 | 3508 | <env name="GATOR_DISABLE">0</env> |
3494 | | - <env name="GPU_TILE_COMPACT">/soft/tools/mpi_wrapper_utils/gpu_tile_compact.sh</env> |
| 3509 | + <env name="GPU_TILE_COMPACT">/lus/flare/projects/E3SM_Dec/tools/mpi_wrapper_utils/gpu_tile_compact.sh</env> |
3495 | 3510 | <env name="FI_CXI_DEFAULT_CQ_SIZE">131072</env> |
3496 | 3511 | <env name="FI_CXI_CQ_FILL_PERCENT">20</env> |
3497 | | - <env name="Kokkos_ROOT">$ENV{KOKKOS_ROOT}</env> |
| 3512 | + <!--<env name="Kokkos_ROOT">$ENV{KOKKOS_ROOT}</env>--> |
3498 | 3513 | <env name="ZES_ENABLE_SYSMAN">1</env> |
3499 | | - <env name="ZEX_NUMBER_OF_CCS">0:4,1:4,2:4,3:4:4:4,5:4,6:4,7:4,8:4,9:4,10:4,11:4</env> |
| 3514 | + <!--<env name="ZEX_NUMBER_OF_CCS">0:4,1:4,2:4,3:4:4:4,5:4,6:4,7:4,8:4,9:4,10:4,11:4</env>--> |
3500 | 3515 | </environment_variables> |
3501 | 3516 | <environment_variables compiler="oneapi-ifx"> |
3502 | 3517 | <env name="LIBOMPTARGET_DEBUG">0</env><!--default 0, max 5 --> |
|
0 commit comments