#include <perf_metrics.h>
◆ num_cpus
◆ num_mpi_ranks
◆ num_nodes
◆ avg_cpus
◆ num_gpus
◆ cycles
◆ instructions
◆ num_measurements
◆ num_mpi_calls
◆ num_omp_parallels
| int64_t num_omp_parallels |
◆ num_omp_tasks
◆ num_gpu_runtime_calls
| int64_t num_gpu_runtime_calls |
◆ elapsed_time
◆ useful_time
◆ mpi_time
◆ omp_load_imbalance_time
| int64_t omp_load_imbalance_time |
◆ omp_scheduling_time
| int64_t omp_scheduling_time |
◆ omp_serialization_time
| int64_t omp_serialization_time |
◆ gpu_runtime_time
◆ min_mpi_normd_proc
| double min_mpi_normd_proc |
MPI time normalized at process level of the process with less MPI i.e.: min(normalize_proc(mpi_times[]))
◆ min_mpi_normd_node
| double min_mpi_normd_node |
MPI time normalized at node level of the node with less MPI i.e.: min(normalize_node(mpi_times[]))
◆ gpu_useful_time
◆ gpu_communication_time
| int64_t gpu_communication_time |
◆ gpu_inactive_time
| int64_t gpu_inactive_time |
◆ max_gpu_useful_time
| int64_t max_gpu_useful_time |
◆ max_gpu_active_time
| int64_t max_gpu_active_time |
The documentation for this struct was generated from the following file: