Something looks fishy about your numbers. The first two sets of numbers
look the same and the last set do look better for the most part. Your
mpirun command line looks weird to me with the "-mca
orte_base_help_aggregate btl,openib,self," did something get chopped off
with the text copy? You should have had a "-mca btl openib,self". Can
you do a run with "-mca btl tcp,self", it should be slower.
I really wouldn't have expected another compiler over IB to be that
dramatically lower performing.
--td
On 04/06/2011 12:40 PM, Nehemiah Dacres wrote:
also, I'm not sure if I'm reading the results right. According to the
last run, did using the sun compilers (update 1 ) result in higher
performance with sunct?
On Wed, Apr 6, 2011 at 11:38 AM, Nehemiah Dacres <dacre...@slu.edu
<mailto:dacre...@slu.edu>> wrote:
some tests I did. I hope this isn't an abuse of the list. please
tell me if it is but thanks to all those who helped me.
this goes to say that the sun MPI works with programs not
compiled with sun’s compilers.
this first test was run as a base case to see if MPI works., the
sedcond run is to see the speed up using OpenIB provides
jian@therock ~]$ mpirun -machinefile list
/opt/iba/src/mpi_apps/mpi_stress/mpi_stress
Start mpi_stress at Wed Apr 6 10:56:29 2011
Size (bytes) TxMessages TxMillionBytes/s
TxMessages/s
32 10000 2.77
86485.67
64 10000 5.76
90049.42
128 10000 11.00
85923.85
256 10000 18.78
73344.43
512 10000 34.47
67331.98
1024 10000 34.81
33998.09
2048 10000 17.31
8454.27
4096 10000 18.34
4476.61
8192 10000 25.43
3104.28
16384 10000 15.56
949.50
32768 10000 13.95
425.74
65536 10000 9.88
150.79
131072 8192 11.05
84.31
262144 4096 13.12
50.04
524288 2048 16.54
31.55
1048576 1024 19.92
18.99
2097152 512 22.54
10.75
4194304 256 25.46
6.07
Iteration 0 : errors = 0, total = 0 (495 secs, Wed Apr 6 11:04:44
2011)
After 1 iteration(s), 8 mins and 15 secs, total errors = 0
here is the infiniband run
[jian@therock ~]$ mpirun -mca orte_base_help_aggregate
btl,openib,self, -machinefile list
/opt/iba/src/mpi_apps/mpi_stress/mpi_stress
Start mpi_stress at Wed Apr 6 11:07:06 2011
Size (bytes) TxMessages TxMillionBytes/s
TxMessages/s
32 10000 2.72 84907.69
64 10000 5.83 91097.94
128 10000 10.75 83959.63
256 10000 18.53 72384.48
512 10000 34.96 68285.00
1024 10000 11.40 11133.10
2048 10000 20.88 10196.34
4096 10000 10.13 2472.13
8192 10000 19.32 2358.25
16384 10000 14.58 890.10
32768 10000 15.85 483.61
65536 10000 9.04 137.95
131072 8192 10.90 83.12
262144 4096 13.57
51.76
524288 2048 16.82
32.08
1048576 1024 19.10 18.21
2097152 512 22.13 10.55
4194304 256 21.66 5.16
Iteration 0 : errors = 0, total = 0 (511 secs, Wed Apr 6 11:15:37
2011)
After 1 iteration(s), 8 mins and 31 secs, total errors = 0
compiled with the sun compilers i think
[jian@therock ~]$ mpirun -mca orte_base_help_aggregate
btl,openib,self, -machinefile list sunMpiStress
Start mpi_stress at Wed Apr 6 11:23:18 2011
Size (bytes) TxMessages TxMillionBytes/s
TxMessages/s
32 10000 2.60
81159.60
64 10000 5.19
81016.95
128 10000 10.23
79953.34
256 10000 16.74
65406.52
512 10000 23.71
46304.92
1024 10000 54.62
53340.73
2048 10000 45.75
22340.58
4096 10000 29.32
7158.87
8192 10000 28.61
3492.77
16384 10000 184.03
11232.26
32768 10000 215.69
6582.21
65536 10000 229.88
3507.64
131072 8192 231.64
1767.25
262144 4096 220.73
842.00
524288 2048 121.61
231.95
1048576 1024 66.54
63.46
2097152 512 44.20
21.08
4194304 256 45.17
10.77
Iteration 0 : errors = 0, total = 0 (93 secs, Wed Apr 6 11:24:52
2011)
After 1 iteration(s), 1 mins and 33 secs, total errors = 0
sanity check: was sunMpiStress compiled using the sun compilers or
oracle compilerrs ?
[jian@therock ~]$ which mpirun
/opt/SUNWhpc/HPC8.2.1c/sun/bin/mpirun
[jian@therock ~]$ ldd sunMpiStress
libmpi.so.0 => /opt/SUNWhpc/HPC8.2.1c/sun/lib/lib64/libmpi.so.0
(0x00002b5d2c6c3000)
libopen-rte.so.0 =>
/opt/SUNWhpc/HPC8.2.1c/sun/lib/lib64/libopen-rte.so.0
(0x00002b5d2c8c1000)
libopen-pal.so.0 =>
/opt/SUNWhpc/HPC8.2.1c/sun/lib/lib64/libopen-pal.so.0
(0x00002b5d2ca19000)
libnsl.so.1 => /lib64/libnsl.so.1 (0x0000003361400000)
librt.so.1 => /lib64/librt.so.1 (0x000000335f400000)
libm.so.6 => /lib64/libm.so.6 (0x000000335e400000)
libdl.so.2 => /lib64/libdl.so.2 (0x000000335e800000)
libutil.so.1 => /lib64/libutil.so.1 (0x000000336ba00000)
libpthread.so.0 => /lib64/libpthread.so.0 (0x000000335ec00000)
libc.so.6 => /lib64/libc.so.6 (0x000000335e000000)
/lib64/ld-linux-x86-64.so.2 (0x000000335dc00000)
[jian@therock ~]$ which mpicc
/opt/SUNWhpc/HPC8.2.1c/sun/bin/mpicc
[jian@therock ~]$ mpicc
/opt/iba/src/mpi_apps/mpi_stress/mpi_stress.c -o sunMpiStress --showme
cc/opt/iba/src/mpi_apps/mpi_stress/mpi_stress.c -o sunMpiStress
-I/opt/SUNWhpc/HPC8.2.1c/sun/include/64
-I/opt/SUNWhpc/HPC8.2.1c/sun/include/64/openmpi
-R/opt/mx/lib/lib64 -R/opt/SUNWhpc/HPC8.2.1c/sun/lib/lib64
-L/opt/SUNWhpc/HPC8.2.1c/sun/lib/lib64 -lmpi -lopen-rte -lopen-pal
-lnsl -lrt -lm -ldl -lutil -lpthread
[jian@therock ~]$ which cc
/opt/sun/sunstudio12.1/bin/cc
looks like it!
--
Nehemiah I. Dacres
System Administrator
Advanced Technology Group Saint Louis University
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
--
Oracle
Terry D. Dontje | Principal Software Engineer
Developer Tools Engineering | +1.781.442.2631
Oracle *- Performance Technologies*
95 Network Drive, Burlington, MA 01803
Email terry.don...@oracle.com <mailto:terry.don...@oracle.com>
