diff options
| author | Bill Wendling <wendling@ncsa.uiuc.edu> | 2002-01-30 04:16:15 (GMT) |
|---|---|---|
| committer | Bill Wendling <wendling@ncsa.uiuc.edu> | 2002-01-30 04:16:15 (GMT) |
| commit | cb905d215efc4c23eda2fc23271384f13c60bb36 (patch) | |
| tree | f94a22211cb9219ccd0a77ca274d23c2ecf34abd /perform/pio_perf.h | |
| parent | 61f6b93eb736e989b21a8cddfae0bb45052df6ea (diff) | |
| download | hdf5-cb905d215efc4c23eda2fc23271384f13c60bb36.zip hdf5-cb905d215efc4c23eda2fc23271384f13c60bb36.tar.gz hdf5-cb905d215efc4c23eda2fc23271384f13c60bb36.tar.bz2 | |
[svn-r4882] Purpose:
Feature add and algorithm reworking.
Description:
Added a "--debug" flag so we can print out various extra debugging
information.
Reworked the algorithm so that it's printing the correct throughput.
Here's how it's supposed to work:
process
T_0 T_1 T_2 T_3 ... T_n
iteration 1
2
.
.
.
m
Retrieve the maximum time from each iteration over the number of
processes. (So, if T_i had the maximum time in iteration j, then use
that time). Calculate the "Throughput" of iteration j:
S_j = (raw_size / T_i)
Collect that information over all of the iterations. Then output the
Max, Min, and Ave of all of the S_k's.
Platforms tested:
Linux (pp)
Diffstat (limited to 'perform/pio_perf.h')
| -rw-r--r-- | perform/pio_perf.h | 8 |
1 files changed, 7 insertions, 1 deletions
diff --git a/perform/pio_perf.h b/perform/pio_perf.h index 5ae3faa..cf84f55 100644 --- a/perform/pio_perf.h +++ b/perform/pio_perf.h @@ -41,7 +41,13 @@ typedef struct results_ { extern MPI_Comm pio_comm_g; /* Communicator to run the PIO */ extern int pio_mpi_rank_g; /* MPI rank of pio_comm_g */ -extern int pio_mpi_nprocs_g; /* number of processes of pio_comm_g */ +extern int pio_mpi_nprocs_g; /* number of processes of pio_comm_g */ +extern int pio_debug_level; /* The debug level: + * 0 - Off + * 1 - Minimal + * 2 - Some more + * 3 - Maximal + */ #ifdef __cplusplus extern "C" { |