Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

From: Scott Weitzenkamp \(sweitzen\) (sweitzen_at_[hidden])
Date: 2006-05-09 12:27:38


Pallas runs OK up to Alltoall test, then we get:

/data/software/qa/MPI/openmpi-1.1a2-rhel4-`uname -m`-mvapi/bin/mpirun
--mca \
mpi_leave_pinned 1 --np 4 --hostfile /tmp/hostfile env
LD_LIBRARY_PATH=/data\
/software/qa/MPI/openmpi-1.1a2-rhel3-`uname
-m`-mvapi/lib:/data/software/qa/\
MPI/openmpi-1.1a2-rhel3-`uname -m`-mvapi/lib64
/data/home/scott/builds/Topsp\
inOS-2.6.0/build183/protest/Igr7/050906_002959/open.pallas/1147159823/PM
B2.2\
.1/SRC_PMB/PMB-MPI1 Alltoall
#---------------------------------------------------^M
# PALLAS MPI Benchmark Suite V2.2, MPI-1 part ^M
#---------------------------------------------------^M
# Date : Tue May 9 00:33:10 2006^M
# Machine : i686# System : Linux^M
# Release : 2.4.21-40.ELhugemem^M
# Version : #1 SMP Thu Feb 2 22:14:10 EST 2006^M
^M
#^M
# Minimum message length in bytes: 0^M
# Maximum message length in bytes: 4194304^M
#^M
# MPI_Datatype : MPI_BYTE ^M
# MPI_Datatype for reductions : MPI_FLOAT^M
# MPI_Op : MPI_SUM ^M
#^M
#^M
^M
# List of Benchmarks to run:^M
^M
# Alltoall^M
^M
#----------------------------------------------------------------^M
# Benchmarking Alltoall ^M
# ( #processes = 2 ) ^M
# ( 2 additional processes waiting in MPI_Barrier)^M
#----------------------------------------------------------------^M
       #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]^M
            0 1000 0.03 0.03 0.03^M
            1 1000 8.12 8.18 8.15^M
            2 1000 7.80 7.80 7.80^M
            4 1000 7.74 7.75 7.74^M
            8 1000 7.87 7.88 7.88^M
           16 1000 8.04 8.05 8.04^M
           32 1000 7.96 7.96 7.96^M
           64 1000 9.30 9.31 9.30^M
          128 1000 10.37 10.38 10.38^M
          256 1000 12.03 12.04 12.04^M
          512 1000 12.83 12.85 12.84^M
         1024 1000 16.23 16.25 16.24^M
         2048 1000 20.55 20.57 20.56^M
         4096 1000 26.89 26.89 26.89^M
         8192 1000 44.50 44.51 44.51^M
        16384 1000 78.13 78.18 78.15^M
        32768 1000 132.28 132.31 132.30^M
        65536 640 222.20 222.25 222.23^M
       131072 320 405.07 405.15 405.11^M
       262144 160 1069.56 1069.62 1069.59^M
       524288 80 2479.89 2481.27 2480.58^M
      1048576 40 4875.95 4882.20 4879.08^M
      2097152 20 9520.25 9530.10 9525.18^M
      4194304 10 19020.81 19048.00 19034.41^M
^M
#----------------------------------------------------------------^M
# Benchmarking Alltoall ^M
# ( #processes = 4 ) ^M
#----------------------------------------------------------------^M
       #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]^M
            0 1000 0.03 0.03 0.03^M
            1 1000 27.06 27.07 27.07^M
            2 1000 16.48 16.49 16.48^M
            4 1000 17.28 17.29 17.28^M
            8 1000 20.79 20.80 20.80^M
           16 1000 18.60 18.61 18.61^M
           32 1000 18.46 18.47 18.47^M
           64 1000 19.09 19.10 19.09^M
          128 1000 16.24 16.29 16.26^M
          256 1000 18.10 18.11 18.10^M
          512 1000 19.39 19.40 19.39^M
         1024 1000 26.51 26.52 26.52^M
         2048 1000 37.85 37.86 37.85^M
         4096 1000 60.41 60.42 60.41^M
         8192 1000 103.04 103.08 103.06^M
        16384 1000 200.75 200.79 200.78^M
        32768 1000 2981.05 2982.04 2981.70^M
        65536 640 3435.51 3436.92 3436.32^M
       131072 320 4224.32 4226.30 4225.34^M
       262144 160 6300.70 6305.17 6303.13^M
Signal:11 info.si_errno:0(Success) si_code:1(SEGV_MAPERR)^M
Failing at addr:0x20^M
[0]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel4-i686-mvapi/lib/libopal.so
.0 \
[0xdfd66c]^M
[1] func:/lib/tls/libpthread.so.0 [0x938e40]^M
[2]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_r\
cache_rb.so(mca_rcache_rb_insert+0x30) [0x85012c]^M
[3]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_m\
pool_mvapi.so(mca_mpool_mvapi_register+0x1b3) [0x9d3157]^M
[4]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_p\
ml_ob1.so(mca_pml_ob1_rdma_btls+0x18c) [0x292458]^M
[5]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_p\
ml_ob1.so [0x2951f9]^M
[6]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_p\
ml_ob1.so(mca_pml_ob1_recv_request_progress+0x5e4) [0x294098]^M
[7]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_p\
ml_ob1.so(mca_pml_ob1_recv_request_match_specific+0xb0) [0x2948f0]^M
[8]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_p\
ml_ob1.so(mca_pml_ob1_irecv+0x139) [0x290d01]^M
[9]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_c\
oll_tuned.so(ompi_coll_tuned_sendrecv_actual+0x38) [0x8a0394]^M
[10]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_\
coll_tuned.so(ompi_coll_tuned_alltoall_intra_pairwise+0xb4) [0x8a1550]^M
[11]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/openmpi/mc
a_\
coll_tuned.so(ompi_coll_tuned_alltoall_intra_dec_fixed+0x7d)
[0x8a0531]^M
[12]
func:/data/software/qa/MPI/openmpi-1.1a2-rhel4-i686-mvapi/lib/libmpi.so.
0(\
MPI_Alltoall+0x167) [0xf89daf]^M
[13]
func:/data/home/scott/builds/TopspinOS-2.6.0/build183/protest/Igr7/05090
6_\
002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB-MPI1(Alltoall+0xa4)
[0x804dd\
28]^M
[14]
func:/data/home/scott/builds/TopspinOS-2.6.0/build183/protest/Igr7/05090
6_\
002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB-MPI1(main+0x32e)
[0x804a78a]\
^M
[15] func:/lib/tls/libc.so.6(__libc_start_main+0xf8) [0x126748]^M
[16]
func:/data/home/scott/builds/TopspinOS-2.6.0/build183/protest/Igr7/05090
6_\
002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB-MPI1(free+0x41)
[0x804a3cd]^\
M
*** End of error message ***^M
\300\200Broken pipe^M

Scott Weitzenkamp
SQA and Release Manager
Server Virtualization Business Unit
Cisco Systems
 

> -----Original Message-----
> From: Scott Weitzenkamp (sweitzen)
> Sent: Tuesday, May 09, 2006 12:35 AM
> To: Scott Weitzenkamp (sweitzen)
> Subject: Chassis TopspinOS-2.6.0 and Host Linux-3.2.0 Test
> Results - Total: 1; Pass 0; Fail: 1; Skip: 0
>
> Protest Run Results
>
> Chassis Branch: TopspinOS-2.6.0
> Chassis Build: build183
> IB Hosts: svbu-qa2650-16 svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> IB Hosts OS: host_hugemem_rhel3_40 host_hugemem_rhel3_37
> host_smp_rhel3_40 host_smp_rhel3_37
> Host Branch: Linux-3.2.0
> Host Build: build089
> Topology:
> /data/home/scott/cluster_switch.main/qa/tests/smoke/Igr7.topo
> Logs:
> /data/home/scott/builds/TopspinOS-2.6.0/build183/protest/Igr7/
> 050906_002959
> Logs URL:
> http://svbu-borg.cisco.com/~releng/data/home/scott/builds/Tops
pinOS-2.6.0/build183/protest/Igr7/050906_002959
>
>
> Scenario file: open.pallas.tmp
>
> ###TEST-D: MPI home dir is
> /data/software/qa/MPI/openmpi-1.1a2-rhel3-`uname -m`-mvapi.
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / PingPong
> Test Duration: 3.270 seconds
> Test Performances :
> #---------------------------------------------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #---------------------------------------------------
> #bytes #repetitions t[usec] Mbytes/sec
> 0 1000 10.19 0.00
> 1 1000 7.20 0.13
> 2 1000 7.23 0.26
> 4 1000 7.22 0.53
> 8 1000 7.36 1.04
> 16 1000 7.37 2.07
> 32 1000 7.52 4.06
> 64 1000 8.63 7.07
> 128 1000 8.97 13.61
> 256 1000 9.74 25.06
> 512 1000 10.90 44.80
> 1024 1000 13.51 72.30
> 2048 1000 16.73 116.71
> 4096 1000 22.24 175.65
> 8192 1000 34.19 228.53
> 16384 1000 56.80 275.11
> 32768 1000 82.50 378.79
> 65536 640 127.01 492.09
> 131072 320 218.89 571.07
> 262144 160 394.11 634.34
> 524288 80 748.98 667.58
> 1048576 40 1468.04 681.18
> 2097152 20 2872.90 696.16
> 4194304 10 5738.20 697.08
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / PingPong -multi 1
> Test Duration: 3.169 seconds
> Test Performances :
> #----------------------------------------------------------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #----------------------------------------------------------------
> Group #bytes #repetitions t[usec] Mbytes/sec
>
> 0 0 1000 10.19 0.00
> 1 0 1000 10.22 0.00
>
> 0 1 1000 7.28 0.13
> 1 1 1000 7.13 0.13
>
> 0 2 1000 7.25 0.26
> 1 2 1000 7.12 0.27
>
> 0 4 1000 7.26 0.53
> 1 4 1000 7.15 0.53
>
> 0 8 1000 7.42 1.03
> 1 8 1000 7.27 1.05
>
> 0 16 1000 7.51 2.03
> 1 16 1000 7.24 2.11
>
> 0 32 1000 7.52 4.06
> 1 32 1000 7.51 4.06
>
> 0 64 1000 8.67 7.04
> 1 64 1000 8.73 6.99
>
> 0 128 1000 9.95 12.27
> 1 128 1000 9.02 13.54
>
> 0 256 1000 9.78 24.98
> 1 256 1000 9.71 25.15
>
> 0 512 1000 11.06 44.17
> 1 512 1000 10.80 45.23
>
> 0 1024 1000 13.56 72.02
> 1 1024 1000 13.32 73.31
>
> 0 2048 1000 16.59 117.71
> 1 2048 1000 16.49 118.42
>
> 0 4096 1000 21.96 177.91
> 1 4096 1000 21.50 181.69
>
> 0 8192 1000 32.33 241.67
> 1 8192 1000 33.50 233.23
>
> 0 16384 1000 56.32 277.44
> 1 16384 1000 57.28 272.79
>
> 0 32768 1000 82.25 379.94
> 1 32768 1000 80.67 387.40
>
> 0 65536 640 126.69 493.31
> 1 65536 640 125.46 498.17
>
> 0 131072 320 217.08 575.83
> 1 131072 320 214.56 582.59
>
> 0 262144 160 395.69 631.81
> 1 262144 160 393.47 635.37
>
> 0 524288 80 748.94 667.61
> 1 524288 80 752.60 664.36
>
> 0 1048576 40 1443.01 693.00
> 1 1048576 40 1467.11 681.61
>
> 0 2097152 20 2884.95 693.25
> 1 2097152 20 2909.32 687.44
>
> 0 4194304 10 5721.85 699.07
> 1 4194304 10 5693.45 702.56
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / PingPing
> Test Duration: 8.259 seconds
> Test Performances :
> #---------------------------------------------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #---------------------------------------------------
> #bytes #repetitions t[usec] Mbytes/sec
> 0 1000 13.65 0.00
> 1 1000 7.28 0.13
> 2 1000 7.26 0.26
> 4 1000 7.32 0.52
> 8 1000 7.46 1.02
> 16 1000 7.51 2.03
> 32 1000 7.62 4.01
> 64 1000 8.78 6.95
> 128 1000 9.22 13.24
> 256 1000 10.50 23.25
> 512 1000 11.58 42.17
> 1024 1000 14.83 65.85
> 2048 1000 19.15 101.97
> 4096 1000 26.50 147.42
> 8192 1000 44.17 176.89
> 16384 1000 78.89 198.07
> 32768 1000 123.22 253.62
> 65536 640 215.34 290.23
> 131072 320 366.66 340.91
> 262144 160 688.94 362.88
> 524288 80 1338.12 373.66
> 1048576 40 2629.77 380.26
> 2097152 20 5245.40 381.29
> 4194304 10 10415.61 384.04
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / PingPing -multi 1
> Test Duration: 2.870 seconds
> Test Performances :
> #----------------------------------------------------------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #----------------------------------------------------------------
> Group #bytes #repetitions t[usec] Mbytes/sec
>
> 0 0 1000 13.69 0.00
> 1 0 1000 11.91 0.00
>
> 0 1 1000 7.47 0.13
> 1 1 1000 7.28 0.13
>
> 0 2 1000 7.42 0.26
> 1 2 1000 7.33 0.26
>
> 0 4 1000 7.47 0.51
> 1 4 1000 7.30 0.52
>
> 0 8 1000 7.61 1.00
> 1 8 1000 7.47 1.02
>
> 0 16 1000 7.55 2.02
> 1 16 1000 7.49 2.04
>
> 0 32 1000 7.62 4.00
> 1 32 1000 7.60 4.01
>
> 0 64 1000 8.83 6.91
> 1 64 1000 8.77 6.96
>
> 0 128 1000 11.16 10.94
> 1 128 1000 9.02 13.54
>
> 0 256 1000 10.30 23.71
> 1 256 1000 10.09 24.21
>
> 0 512 1000 12.12 40.28
> 1 512 1000 11.76 41.53
>
> 0 1024 1000 14.96 65.30
> 1 1024 1000 14.95 65.33
>
> 0 2048 1000 19.50 100.15
> 1 2048 1000 19.15 101.99
>
> 0 4096 1000 26.42 147.87
> 1 4096 1000 25.89 150.87
>
> 0 8192 1000 46.07 169.57
> 1 8192 1000 42.60 183.38
>
> 0 16384 1000 79.15 197.41
> 1 16384 1000 76.40 204.52
>
> 0 32768 1000 123.99 252.03
> 1 32768 1000 122.70 254.69
>
> 0 65536 640 204.70 305.32
> 1 65536 640 205.00 304.87
>
> 0 131072 320 370.49 337.39
> 1 131072 320 366.83 340.76
>
> 0 262144 160 692.09 361.23
> 1 262144 160 691.42 361.58
>
> 0 524288 80 1336.15 374.21
> 1 524288 80 1330.53 375.79
>
> 0 1048576 40 2634.95 379.51
> 1 1048576 40 2627.58 380.58
>
> 0 2097152 20 5240.35 381.65
> 1 2097152 20 5194.95 384.99
>
> 0 4194304 10 10409.59 384.26
> 1 4194304 10 10314.59 387.80
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Sendrecv
> Test Duration: 4.710 seconds
> Test Performances :
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #-------------------------------------------------------------
> ----------------
> #bytes #repetitions t_min[usec] t_max[usec]
> t_avg[usec] Mbytes/sec
> 0 1000 13.77 13.78
> 13.78 0.00
> 1 1000 7.37 7.38
> 7.38 0.26
> 2 1000 7.36 7.37
> 7.36 0.52
> 4 1000 7.36 7.36
> 7.36 1.04
> 8 1000 7.52 7.53
> 7.53 2.03
> 16 1000 7.52 7.52
> 7.52 4.06
> 32 1000 7.69 7.70
> 7.70 7.93
> 64 1000 8.80 8.81
> 8.81 13.85
> 128 1000 9.19 9.20
> 9.20 26.54
> 256 1000 10.17 10.17
> 10.17 47.99
> 512 1000 11.77 11.78
> 11.77 82.90
> 1024 1000 15.37 15.37
> 15.37 127.07
> 2048 1000 19.25 19.26
> 19.25 202.79
> 4096 1000 27.31 27.33
> 27.32 285.83
> 8192 1000 42.78 42.81
> 42.79 364.98
> 16384 1000 75.22 75.24
> 75.23 415.35
> 32768 1000 122.82 122.85
> 122.84 508.73
> 65536 640 204.42 204.45
> 204.43 611.41
> 131072 320 366.74 366.81
> 366.78 681.55
> 262144 160 689.38 689.56
> 689.47 725.10
> 524288 80 1335.74 1336.15
> 1335.95 748.42
> 1048576 40 2668.65 2669.30
> 2668.98 749.26
> 2097152 20 5237.05 5238.85
> 5237.95 763.53
> 4194304 10 10393.89 10397.00
> 10395.45 769.45
>
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> ----------------
> #bytes #repetitions t_min[usec] t_max[usec]
> t_avg[usec] Mbytes/sec
> 0 1000 21.71 21.72
> 21.72 0.00
> 1 1000 8.49 8.50
> 8.50 0.22
> 2 1000 8.65 8.66
> 8.65 0.44
> 4 1000 8.55 8.57
> 8.56 0.89
> 8 1000 8.66 8.68
> 8.67 1.76
> 16 1000 8.66 8.70
> 8.69 3.51
> 32 1000 8.86 8.88
> 8.87 6.88
> 64 1000 10.01 10.02
> 10.02 12.18
> 128 1000 10.34 10.35
> 10.34 23.58
> 256 1000 11.35 11.36
> 11.35 43.00
> 512 1000 13.02 13.04
> 13.03 74.91
> 1024 1000 16.90 16.92
> 16.91 115.41
> 2048 1000 19.87 19.90
> 19.89 196.32
> 4096 1000 28.42 28.44
> 28.43 274.75
> 8192 1000 48.98 49.04
> 49.01 318.62
> 16384 1000 90.11 90.21
> 90.15 346.41
> 32768 1000 123.08 123.10
> 123.09 507.74
> 65536 640 204.75 204.80
> 204.77 610.37
> 131072 320 366.26 366.38
> 366.34 682.35
> 262144 160 707.12 707.32
> 707.25 706.89
> 524288 80 1335.55 1335.91
> 1335.72 748.55
> 1048576 40 2634.05 2634.73
> 2634.26 759.09
> 2097152 20 5232.59 5244.75
> 5239.09 762.67
> 4194304 10 10369.80 10408.60
> 10389.80 768.59
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Sendrecv -multi 1
> Test Duration: 4.410 seconds
> Test Performances :
> #-------------------------------------------------------------
> -----------------------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #-------------------------------------------------------------
> -----------------------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec] Mbytes/sec
>
> 0 0 1000 13.72
> 13.72 13.72 0.00
> 1 0 1000 12.41
> 12.41 12.41 0.00
>
> 0 1 1000 7.39
> 7.39 7.39 0.26
> 1 1 1000 7.54
> 7.55 7.54 0.25
>
> 0 2 1000 7.34
> 7.35 7.35 0.52
> 1 2 1000 8.19
> 8.21 8.20 0.46
>
> 0 4 1000 7.34
> 7.35 7.35 1.04
> 1 4 1000 7.48
> 7.51 7.49 1.02
>
> 0 8 1000 7.51
> 7.51 7.51 2.03
> 1 8 1000 7.60
> 7.61 7.61 2.00
>
> 0 16 1000 7.48
> 7.48 7.48 4.08
> 1 16 1000 7.67
> 7.67 7.67 3.98
>
> 0 32 1000 7.62
> 7.63 7.63 8.00
> 1 32 1000 7.85
> 7.85 7.85 7.77
>
> 0 64 1000 8.76
> 8.77 8.76 13.92
> 1 64 1000 8.90
> 8.90 8.90 13.71
>
> 0 128 1000 9.27
> 9.28 9.27 26.32
> 1 128 1000 9.45
> 9.46 9.46 25.81
>
> 0 256 1000 10.23
> 10.24 10.24 47.68
> 1 256 1000 10.34
> 10.34 10.34 47.20
>
> 0 512 1000 11.81
> 11.82 11.81 82.64
> 1 512 1000 11.99
> 11.99 11.99 81.46
>
> 0 1024 1000 15.29
> 15.30 15.30 127.64
> 1 1024 1000 15.24
> 15.24 15.24 128.17
>
> 0 2048 1000 19.18
> 19.19 19.19 203.60
> 1 2048 1000 19.38
> 19.38 19.38 201.53
>
> 0 4096 1000 26.37
> 26.38 26.37 296.14
> 1 4096 1000 26.06
> 26.07 26.07 299.65
>
> 0 8192 1000 44.89
> 44.94 44.92 347.69
> 1 8192 1000 42.71
> 42.73 42.72 365.69
>
> 0 16384 1000 78.07
> 78.12 78.09 400.04
> 1 16384 1000 78.14
> 78.17 78.15 399.79
>
> 0 32768 1000 122.95
> 122.99 122.97 508.19
> 1 32768 1000 122.67
> 122.68 122.67 509.47
>
> 0 65536 640 204.46
> 204.51 204.49 611.22
> 1 65536 640 203.90
> 203.93 203.92 612.96
>
> 0 131072 320 366.34
> 366.44 366.39 682.25
> 1 131072 320 366.57
> 366.61 366.59 681.92
>
> 0 262144 160 695.71
> 695.89 695.80 718.50
> 1 262144 160 690.64
> 690.73 690.69 723.87
>
> 0 524288 80 1337.91
> 1338.30 1338.11 747.22
> 1 524288 80 1335.54
> 1335.66 1335.60 748.69
>
> 0 1048576 40 2635.45
> 2636.05 2635.75 758.71
> 1 1048576 40 2624.10
> 2624.18 2624.14 762.14
>
> 0 2097152 20 5242.25
> 5243.81 5243.03 762.80
> 1 2097152 20 5176.90
> 5177.15 5177.03 772.63
>
> 0 4194304 10 10407.00
> 10410.61 10408.80 768.45
> 1 4194304 10 10287.11
> 10287.41 10287.26 777.65
>
> #-------------------------------------------------------------
> -----------------------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> -----------------------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec] Mbytes/sec
>
> 0 0 1000 18.80
> 18.81 18.81 0.00
>
> 0 1 1000 8.76
> 8.77 8.76 0.22
>
> 0 2 1000 8.77
> 8.79 8.78 0.43
>
> 0 4 1000 8.66
> 8.69 8.67 0.88
>
> 0 8 1000 8.84
> 8.85 8.85 1.72
>
> 0 16 1000 8.88
> 8.89 8.89 3.43
>
> 0 32 1000 9.03
> 9.11 9.08 6.70
>
> 0 64 1000 10.09
> 10.10 10.10 12.08
>
> 0 128 1000 11.28
> 11.31 11.29 21.59
>
> 0 256 1000 11.60
> 11.62 11.61 42.02
>
> 0 512 1000 12.97
> 12.98 12.98 75.22
>
> 0 1024 1000 16.27
> 16.29 16.28 119.91
>
> 0 2048 1000 20.33
> 20.37 20.35 191.78
>
> 0 4096 1000 27.75
> 27.77 27.76 281.36
>
> 0 8192 1000 53.92
> 53.97 53.95 289.50
>
> 0 16384 1000 94.15
> 94.19 94.17 331.77
>
> 0 32768 1000 123.83
> 123.84 123.84 504.67
>
> 0 65536 640 204.36
> 204.40 204.39 611.54
>
> 0 131072 320 370.60
> 370.69 370.65 674.42
>
> 0 262144 160 693.22
> 693.38 693.33 721.10
>
> 0 524288 80 1341.00
> 1341.40 1341.18 745.49
>
> 0 1048576 40 2637.87
> 2640.20 2639.39 757.52
>
> 0 2097152 20 5228.25
> 5233.15 5230.40 764.36
>
> 0 4194304 10 10389.59
> 10425.91 10408.58 767.32
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Exchange
> Test Duration: 6.329 seconds
> Test Performances :
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #-------------------------------------------------------------
> ----------------
> #bytes #repetitions t_min[usec] t_max[usec]
> t_avg[usec] Mbytes/sec
> 0 1000 10.01 11.46
> 10.74 0.00
> 1 1000 10.02 10.02
> 10.02 0.38
> 2 1000 9.90 9.91
> 9.91 0.77
> 4 1000 9.93 9.93
> 9.93 1.54
> 8 1000 9.91 9.91
> 9.91 3.08
> 16 1000 10.02 10.02
> 10.02 6.09
> 32 1000 10.18 10.18
> 10.18 11.99
> 64 1000 11.34 11.34
> 11.34 21.53
> 128 1000 11.46 11.46
> 11.46 42.59
> 256 1000 13.09 13.09
> 13.09 74.59
> 512 1000 14.22 14.23
> 14.23 137.27
> 1024 1000 21.13 21.15
> 21.14 184.68
> 2048 1000 27.45 27.47
> 27.46 284.40
> 4096 1000 41.29 41.29
> 41.29 378.40
> 8192 1000 70.92 70.96
> 70.94 440.39
> 16384 1000 129.32 129.32
> 129.32 483.28
> 32768 1000 243.38 243.43
> 243.41 513.49
> 65536 640 402.20 402.27
> 402.23 621.48
> 131072 320 728.52 728.68
> 728.60 686.17
> 262144 160 1370.86 1371.14
> 1371.00 729.32
> 524288 80 2664.69 2665.32
> 2665.01 750.38
> 1048576 40 5262.72 5264.00
> 5263.36 759.88
> 2097152 20 10436.70 10439.55
> 10438.13 766.32
> 4194304 10 20823.30 20828.69
> 20825.99 768.17
>
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> ----------------
> #bytes #repetitions t_min[usec] t_max[usec]
> t_avg[usec] Mbytes/sec
> 0 1000 18.99 19.01
> 19.00 0.00
> 1 1000 10.33 10.34
> 10.33 0.37
> 2 1000 10.26 10.27
> 10.26 0.74
> 4 1000 10.53 10.54
> 10.53 1.45
> 8 1000 10.44 10.45
> 10.45 2.92
> 16 1000 10.62 10.63
> 10.63 5.74
> 32 1000 10.62 10.63
> 10.63 11.48
> 64 1000 12.11 12.12
> 12.12 20.14
> 128 1000 12.18 12.18
> 12.18 40.08
> 256 1000 13.80 13.81
> 13.81 70.71
> 512 1000 15.60 15.61
> 15.60 125.13
> 1024 1000 21.15 21.16
> 21.15 184.60
> 2048 1000 28.88 28.90
> 28.89 270.37
> 4096 1000 42.20 42.23
> 42.22 370.02
> 8192 1000 71.28 71.32
> 71.30 438.18
> 16384 1000 133.44 133.47
> 133.45 468.28
> 32768 1000 237.90 237.94
> 237.93 525.33
> 65536 640 401.07 401.13
> 401.10 623.23
> 131072 320 729.60 729.73
> 729.69 685.19
> 262144 160 1402.19 1402.44
> 1402.34 713.04
> 524288 80 2665.29 2665.88
> 2665.54 750.22
> 1048576 40 5250.57 5251.60
> 5251.06 761.67
> 2097152 20 10444.26 10447.40
> 10445.36 765.74
> 4194304 10 20819.70 20832.89
> 20825.37 768.02
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Exchange -multi 1
> Test Duration: 6.289 seconds
> Test Performances :
> #-------------------------------------------------------------
> -----------------------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #-------------------------------------------------------------
> -----------------------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec] Mbytes/sec
>
> 0 0 1000 9.97
> 11.54 10.76 0.00
> 1 0 1000 13.24
> 13.25 13.25 0.00
>
> 0 1 1000 9.97
> 9.97 9.97 0.38
> 1 1 1000 9.61
> 9.61 9.61 0.40
>
> 0 2 1000 9.91
> 9.92 9.92 0.77
> 1 2 1000 9.52
> 9.53 9.53 0.80
>
> 0 4 1000 9.96
> 9.96 9.96 1.53
> 1 4 1000 9.63
> 9.63 9.63 1.58
>
> 0 8 1000 9.97
> 9.98 9.98 3.06
> 1 8 1000 9.59
> 9.61 9.60 3.18
>
> 0 16 1000 9.95
> 9.96 9.95 6.13
> 1 16 1000 9.81
> 9.81 9.81 6.22
>
> 0 32 1000 10.16
> 10.17 10.16 12.01
> 1 32 1000 9.84
> 9.85 9.85 12.39
>
> 0 64 1000 11.03
> 11.04 11.03 22.12
> 1 64 1000 10.79
> 10.80 10.79 22.61
>
> 0 128 1000 13.15
> 13.15 13.15 37.13
> 1 128 1000 11.15
> 11.16 11.16 43.75
>
> 0 256 1000 12.71
> 12.71 12.71 76.85
> 1 256 1000 12.36
> 12.37 12.36 78.97
>
> 0 512 1000 14.65
> 14.65 14.65 133.33
> 1 512 1000 14.91
> 14.92 14.92 130.94
>
> 0 1024 1000 20.85
> 20.86 20.86 187.28
> 1 1024 1000 20.43
> 20.45 20.44 191.02
>
> 0 2048 1000 27.83
> 27.84 27.84 280.61
> 1 2048 1000 26.73
> 26.75 26.74 292.09
>
> 0 4096 1000 41.41
> 41.43 41.42 377.11
> 1 4096 1000 39.94
> 39.96 39.95 390.99
>
> 0 8192 1000 75.95
> 76.00 75.98 411.20
> 1 8192 1000 68.93
> 68.96 68.94 453.19
>
> 0 16384 1000 132.65
> 132.72 132.69 470.91
> 1 16384 1000 128.51
> 128.55 128.53 486.18
>
> 0 32768 1000 242.75
> 242.79 242.77 514.84
> 1 32768 1000 238.97
> 238.98 238.98 523.05
>
> 0 65536 640 404.87
> 404.95 404.91 617.36
> 1 65536 640 399.60
> 399.62 399.61 625.60
>
> 0 131072 320 728.02
> 728.17 728.10 686.65
> 1 131072 320 724.31
> 724.36 724.33 690.27
>
> 0 262144 160 1369.27
> 1369.57 1369.42 730.16
> 1 262144 160 1365.21
> 1365.30 1365.25 732.44
>
> 0 524288 80 2666.80
> 2667.48 2667.14 749.77
> 1 524288 80 2652.58
> 2652.63 2652.60 753.97
>
> 0 1048576 40 5247.70
> 5248.80 5248.25 762.08
> 1 1048576 40 5220.70
> 5221.15 5220.93 766.11
>
> 0 2097152 20 10438.40
> 10440.86 10439.63 766.22
> 1 2097152 20 10409.95
> 10410.95 10410.45 768.42
>
> 0 4194304 10 20799.19
> 20804.09 20801.64 769.08
> 1 4194304 10 20591.10
> 20591.31 20591.21 777.03
>
> #-------------------------------------------------------------
> -----------------------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> -----------------------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec] Mbytes/sec
>
> 0 0 1000 16.83
> 16.85 16.84 0.00
>
> 0 1 1000 10.28
> 10.29 10.28 0.37
>
> 0 2 1000 10.27
> 10.28 10.27 0.74
>
> 0 4 1000 10.27
> 10.28 10.28 1.48
>
> 0 8 1000 10.35
> 10.38 10.36 2.94
>
> 0 16 1000 10.40
> 10.42 10.41 5.86
>
> 0 32 1000 10.59
> 10.61 10.60 11.51
>
> 0 64 1000 11.73
> 11.74 11.74 20.79
>
> 0 128 1000 12.19
> 12.20 12.19 40.03
>
> 0 256 1000 13.37
> 13.38 13.38 72.97
>
> 0 512 1000 15.58
> 15.59 15.59 125.25
>
> 0 1024 1000 21.11
> 21.13 21.12 184.85
>
> 0 2048 1000 28.09
> 28.10 28.10 277.99
>
> 0 4096 1000 41.51
> 41.54 41.52 376.17
>
> 0 8192 1000 69.64
> 69.68 69.66 448.50
>
> 0 16384 1000 129.73
> 129.79 129.76 481.57
>
> 0 32768 1000 240.58
> 240.64 240.62 519.45
>
> 0 65536 640 403.15
> 403.23 403.20 620.00
>
> 0 131072 320 733.58
> 733.76 733.70 681.42
>
> 0 262144 160 1438.44
> 1438.87 1438.72 694.99
>
> 0 524288 80 2671.47
> 2672.16 2671.92 748.46
>
> 0 1048576 40 5282.70
> 5284.03 5283.49 757.00
>
> 0 2097152 20 10453.80
> 10461.55 10458.58 764.71
>
> 0 4194304 10 20774.30
> 20786.60 20779.53 769.73
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Reduce
> Test Duration: 6.999 seconds
> Test Performances :
> #----------------------------------------------------------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.07 0.09 0.08
> 4 1000 15.09 15.10 15.10
> 8 1000 7.82 7.82 7.82
> 16 1000 7.63 7.63 7.63
> 32 1000 7.96 7.96 7.96
> 64 1000 9.00 9.00 9.00
> 128 1000 9.41 9.41 9.41
> 256 1000 10.39 10.39 10.39
> 512 1000 11.66 11.66 11.66
> 1024 1000 14.66 14.67 14.66
> 2048 1000 18.60 18.60 18.60
> 4096 1000 25.24 25.25 25.25
> 8192 1000 39.60 39.62 39.61
> 16384 1000 67.17 67.21 67.19
> 32768 1000 110.43 110.43 110.43
> 65536 640 179.86 179.90 179.88
> 131072 320 328.67 328.93 328.80
> 262144 160 1437.36 1441.53 1439.44
> 524288 80 2702.94 2718.30 2710.62
> 1048576 40 5261.58 5317.90 5289.74
> 2097152 20 10345.46 10573.29 10459.37
> 4194304 10 20220.49 21127.30 20673.90
>
> #----------------------------------------------------------------
> # ( #processes = 4 )
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.07 0.07 0.07
> 4 1000 95.97 96.01 95.99
> 8 1000 8.92 8.94 8.93
> 16 1000 8.89 8.91 8.90
> 32 1000 9.11 9.19 9.16
> 64 1000 10.32 10.34 10.33
> 128 1000 10.92 10.95 10.93
> 256 1000 12.24 12.27 12.25
> 512 1000 13.94 13.98 13.96
> 1024 1000 17.59 17.63 17.61
> 2048 1000 23.05 23.11 23.07
> 4096 1000 33.54 33.61 33.57
> 8192 1000 54.91 55.04 54.97
> 16384 1000 105.20 105.43 105.30
> 32768 1000 282.82 283.05 282.94
> 65536 640 473.08 473.68 473.41
> 131072 320 872.08 874.50 873.42
> 262144 160 2486.53 2498.51 2493.06
> 524288 80 5017.72 5074.60 5048.90
> 1048576 40 11351.25 11573.85 11467.14
> 2097152 20 22447.00 23285.15 22888.91
> 4194304 10 41543.21 46419.71 44247.70
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Reduce -multi 1
> Test Duration: 7.000 seconds
> Test Performances :
> #-------------------------------------------------------------
> ----------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.07
> 0.07 0.07
> 1 0 1000 0.06
> 0.06 0.06
>
> 0 4 1000 14.91
> 14.92 14.91
> 1 4 1000 15.19
> 15.20 15.19
>
> 0 8 1000 7.63
> 7.63 7.63
> 1 8 1000 7.53
> 7.54 7.53
>
> 0 16 1000 7.62
> 7.62 7.62
> 1 16 1000 7.61
> 7.62 7.62
>
> 0 32 1000 7.74
> 7.75 7.75
> 1 32 1000 7.76
> 7.77 7.77
>
> 0 64 1000 8.94
> 8.95 8.94
> 1 64 1000 8.97
> 8.98 8.98
>
> 0 128 1000 9.38
> 9.38 9.38
> 1 128 1000 9.45
> 9.46 9.46
>
> 0 256 1000 10.34
> 10.34 10.34
> 1 256 1000 10.17
> 10.18 10.18
>
> 0 512 1000 13.35
> 13.35 13.35
> 1 512 1000 11.46
> 11.47 11.47
>
> 0 1024 1000 14.54
> 14.55 14.54
> 1 1024 1000 14.41
> 14.43 14.42
>
> 0 2048 1000 18.28
> 18.29 18.29
> 1 2048 1000 18.30
> 18.32 18.31
>
> 0 4096 1000 25.36
> 25.38 25.37
> 1 4096 1000 25.11
> 25.14 25.13
>
> 0 8192 1000 39.01
> 39.04 39.02
> 1 8192 1000 40.01
> 40.05 40.03
>
> 0 16384 1000 67.16
> 67.20 67.18
> 1 16384 1000 68.11
> 68.18 68.14
>
> 0 32768 1000 109.07
> 109.08 109.07
> 1 32768 1000 108.65
> 108.69 108.67
>
> 0 65536 640 181.28
> 181.33 181.30
> 1 65536 640 179.40
> 179.49 179.44
>
> 0 131072 320 329.42
> 329.68 329.55
> 1 131072 320 327.54
> 327.90 327.72
>
> 0 262144 160 1441.79
> 1446.28 1444.03
> 1 262144 160 1399.98
> 1404.23 1402.10
>
> 0 524288 80 2741.36
> 2757.04 2749.20
> 1 524288 80 2698.99
> 2714.05 2706.52
>
> 0 1048576 40 5220.40
> 5278.08 5249.24
> 1 1048576 40 5121.60
> 5177.47 5149.54
>
> 0 2097152 20 10209.80
> 10425.75 10317.78
> 1 2097152 20 10289.00
> 10506.06 10397.53
>
> 0 4194304 10 19867.90
> 20674.00 20270.95
> 1 4194304 10 19935.30
> 20791.00 20363.15
>
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.07
> 0.07 0.07
>
> 0 4 1000 18.01
> 18.05 18.03
>
> 0 8 1000 9.03
> 9.05 9.04
>
> 0 16 1000 9.01
> 9.04 9.03
>
> 0 32 1000 9.23
> 9.26 9.25
>
> 0 64 1000 10.35
> 10.39 10.37
>
> 0 128 1000 10.98
> 11.02 11.00
>
> 0 256 1000 12.33
> 12.37 12.35
>
> 0 512 1000 13.84
> 13.89 13.86
>
> 0 1024 1000 17.61
> 17.66 17.64
>
> 0 2048 1000 23.02
> 23.09 23.05
>
> 0 4096 1000 33.41
> 33.49 33.45
>
> 0 8192 1000 56.72
> 56.86 56.78
>
> 0 16384 1000 105.83
> 106.06 105.94
>
> 0 32768 1000 284.02
> 284.25 284.14
>
> 0 65536 640 478.39
> 479.01 478.73
>
> 0 131072 320 871.73
> 874.10 873.04
>
> 0 262144 160 2718.90
> 2733.26 2726.07
>
> 0 524288 80 5462.75
> 5521.28 5492.48
>
> 0 1048576 40 11273.95
> 11497.07 11389.24
>
> 0 2097152 20 22380.00
> 23276.70 22850.69
>
> 0 4194304 10 41432.80
> 46174.20 44063.35
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Reduce_scatter
> Test Duration: 13.139 seconds
> Test Performances :
> #----------------------------------------------------------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.09 0.09 0.09
> 4 1000 103.56 110.11 106.83
> 8 1000 15.40 15.41 15.41
> 16 1000 15.46 15.46 15.46
> 32 1000 15.58 15.58 15.58
> 64 1000 16.83 16.83 16.83
> 128 1000 18.28 18.28 18.28
> 256 1000 19.74 19.74 19.74
> 512 1000 21.85 21.85 21.85
> 1024 1000 26.03 26.03 26.03
> 2048 1000 32.28 32.28 32.28
> 4096 1000 42.88 42.88 42.88
> 8192 1000 62.70 62.70 62.70
> 16384 1000 102.41 102.42 102.41
> 32768 1000 170.04 170.05 170.05
> 65536 640 276.37 276.41 276.39
> 131072 320 577.86 578.05 577.96
> 262144 160 2278.61 2281.04 2279.83
> 524288 80 4393.82 4399.40 4396.61
> 1048576 40 8482.20 8497.58 8489.89
> 2097152 20 16983.75 17031.95 17007.85
> 4194304 10 33703.21 33863.61 33783.41
>
> #----------------------------------------------------------------
> # ( #processes = 4 )
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.16 0.17 0.16
> 4 1000 1.58 125.57 82.74
> 8 1000 1.63 23.59 12.62
> 16 1000 26.44 26.46 26.45
> 32 1000 26.57 26.59 26.58
> 64 1000 27.85 27.87 27.86
> 128 1000 28.60 28.62 28.61
> 256 1000 28.45 28.47 28.46
> 512 1000 33.40 33.42 33.41
> 1024 1000 34.57 34.59 34.58
> 2048 1000 44.37 44.39 44.38
> 4096 1000 63.89 63.91 63.91
> 8192 1000 101.82 101.85 101.84
> 16384 1000 174.28 174.33 174.31
> 32768 1000 1344.88 1345.14 1345.07
> 65536 640 673.66 673.87 673.79
> 131072 320 4261.23 4266.19 4264.25
> 262144 160 6760.22 6771.17 6766.40
> 524288 80 11947.40 11974.67 11962.54
> 1048576 40 22110.45 22185.45 22152.57
> 2097152 20 36873.10 37003.40 36937.47
> 4194304 10 72581.01 73049.29 72814.97
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Reduce_scatter -multi 1
> Test Duration: 13.139 seconds
> Test Performances :
> #-------------------------------------------------------------
> ----------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.09
> 0.09 0.09
> 1 0 1000 0.08
> 0.08 0.08
>
> 0 4 1000 104.08
> 107.45 105.77
> 1 4 1000 103.74
> 112.48 108.11
>
> 0 8 1000 15.12
> 15.13 15.12
> 1 8 1000 15.06
> 15.07 15.06
>
> 0 16 1000 15.16
> 15.17 15.17
> 1 16 1000 14.90
> 14.91 14.91
>
> 0 32 1000 15.43
> 15.44 15.43
> 1 32 1000 15.11
> 15.12 15.11
>
> 0 64 1000 16.80
> 16.81 16.81
> 1 64 1000 16.60
> 16.61 16.61
>
> 0 128 1000 18.23
> 18.23 18.23
> 1 128 1000 17.93
> 17.94 17.93
>
> 0 256 1000 20.14
> 20.14 20.14
> 1 256 1000 19.08
> 19.09 19.09
>
> 0 512 1000 23.24
> 23.24 23.24
> 1 512 1000 21.20
> 21.21 21.20
>
> 0 1024 1000 25.82
> 25.83 25.82
> 1 1024 1000 25.32
> 25.32 25.32
>
> 0 2048 1000 31.92
> 31.92 31.92
> 1 2048 1000 31.67
> 31.69 31.68
>
> 0 4096 1000 42.01
> 42.01 42.01
> 1 4096 1000 41.96
> 41.98 41.97
>
> 0 8192 1000 62.32
> 62.32 62.32
> 1 8192 1000 61.64
> 61.67 61.65
>
> 0 16384 1000 101.94
> 101.95 101.94
> 1 16384 1000 101.36
> 101.39 101.37
>
> 0 32768 1000 165.90
> 165.92 165.91
> 1 32768 1000 167.94
> 167.99 167.96
>
> 0 65536 640 279.32
> 279.37 279.35
> 1 65536 640 276.40
> 276.40 276.40
>
> 0 131072 320 583.91
> 584.11 584.01
> 1 131072 320 530.21
> 530.24 530.22
>
> 0 262144 160 2297.08
> 2299.36 2298.22
> 1 262144 160 2271.50
> 2273.19 2272.35
>
> 0 524288 80 4494.29
> 4500.10 4497.19
> 1 524288 80 4328.51
> 4333.54 4331.03
>
> 0 1048576 40 8386.25
> 8401.35 8393.80
> 1 1048576 40 8607.18
> 8620.70 8613.94
>
> 0 2097152 20 17036.05
> 17084.65 17060.35
> 1 2097152 20 17381.00
> 17426.40 17403.70
>
> 0 4194304 10 33299.60
> 33459.90 33379.75
> 1 4194304 10 34933.89
> 35096.40 35015.14
>
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.16
> 0.17 0.16
>
> 0 4 1000 1.61
> 115.35 57.34
>
> 0 8 1000 1.60
> 23.64 12.62
>
> 0 16 1000 27.11
> 27.13 27.12
>
> 0 32 1000 27.13
> 27.15 27.14
>
> 0 64 1000 28.15
> 28.17 28.16
>
> 0 128 1000 29.04
> 29.06 29.05
>
> 0 256 1000 30.94
> 30.96 30.95
>
> 0 512 1000 33.68
> 33.70 33.69
>
> 0 1024 1000 36.91
> 36.93 36.92
>
> 0 2048 1000 45.90
> 45.91 45.91
>
> 0 4096 1000 65.67
> 65.69 65.68
>
> 0 8192 1000 103.55
> 103.58 103.57
>
> 0 16384 1000 177.91
> 177.96 177.94
>
> 0 32768 1000 1353.36
> 1353.63 1353.55
>
> 0 65536 640 676.91
> 677.09 677.02
>
> 0 131072 320 4241.57
> 4246.66 4244.66
>
> 0 262144 160 6770.91
> 6782.06 6777.22
>
> 0 524288 80 11921.58
> 11948.23 11936.62
>
> 0 1048576 40 22039.17
> 22112.38 22080.08
>
> 0 2097152 20 37074.80
> 37205.05 37139.25
>
> 0 4194304 10 73296.40
> 73763.79 73530.40
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Allreduce
> Test Duration: 9.839 seconds
> Test Performances :
> #----------------------------------------------------------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.04 0.04 0.04
> 4 1000 16.66 18.49 17.58
> 8 1000 15.04 15.04 15.04
> 16 1000 15.08 15.09 15.08
> 32 1000 15.29 15.30 15.30
> 64 1000 17.59 17.59 17.59
> 128 1000 18.43 18.43 18.43
> 256 1000 20.14 20.14 20.14
> 512 1000 22.60 22.60 22.60
> 1024 1000 28.17 28.17 28.17
> 2048 1000 35.94 35.94 35.94
> 4096 1000 46.86 46.86 46.86
> 8192 1000 71.59 71.60 71.59
> 16384 1000 124.06 124.08 124.07
> 32768 1000 200.80 200.84 200.82
> 65536 640 327.94 327.99 327.97
> 131072 320 582.19 582.32 582.25
> 262144 160 1309.89 1310.12 1310.00
> 524288 80 2438.80 2439.33 2439.06
> 1048576 40 4633.92 4634.83 4634.37
> 2097152 20 9253.74 9255.55 9254.65
> 4194304 10 18454.71 18457.90 18456.30
>
> #----------------------------------------------------------------
> # ( #processes = 4 )
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.04 0.05 0.04
> 4 1000 29.43 29.46 29.44
> 8 1000 25.81 25.83 25.82
> 16 1000 26.17 26.19 26.18
> 32 1000 26.22 26.24 26.23
> 64 1000 30.45 30.47 30.46
> 128 1000 31.27 31.29 31.28
> 256 1000 33.91 33.93 33.92
> 512 1000 38.45 38.47 38.46
> 1024 1000 48.97 49.00 48.99
> 2048 1000 59.94 59.97 59.95
> 4096 1000 83.83 83.87 83.85
> 8192 1000 134.24 134.29 134.26
> 16384 1000 245.60 245.77 245.69
> 32768 1000 523.95 524.03 523.98
> 65536 640 881.63 881.78 881.67
> 131072 320 1643.84 1644.43 1644.07
> 262144 160 4608.31 4610.54 4609.24
> 524288 80 9078.54 9087.41 9082.61
> 1048576 40 17319.03 17354.53 17336.08
> 2097152 20 34595.25 34736.90 34664.58
> 4194304 10 68758.90 69328.70 69040.74
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Allreduce -multi 1
> Test Duration: 9.939 seconds
> Test Performances :
> #-------------------------------------------------------------
> ----------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.04
> 0.04 0.04
> 1 0 1000 0.04
> 0.04 0.04
>
> 0 4 1000 15.08
> 16.79 15.94
> 1 4 1000 21.90
> 21.90 21.90
>
> 0 8 1000 14.92
> 14.92 14.92
> 1 8 1000 14.81
> 14.81 14.81
>
> 0 16 1000 15.01
> 15.01 15.01
> 1 16 1000 14.75
> 14.76 14.76
>
> 0 32 1000 15.37
> 15.37 15.37
> 1 32 1000 15.06
> 15.06 15.06
>
> 0 64 1000 17.71
> 17.72 17.71
> 1 64 1000 17.25
> 17.26 17.26
>
> 0 128 1000 18.76
> 18.76 18.76
> 1 128 1000 18.26
> 18.26 18.26
>
> 0 256 1000 20.52
> 20.52 20.52
> 1 256 1000 19.62
> 19.63 19.63
>
> 0 512 1000 24.53
> 24.53 24.53
> 1 512 1000 22.09
> 22.10 22.09
>
> 0 1024 1000 28.33
> 28.33 28.33
> 1 1024 1000 27.53
> 27.54 27.54
>
> 0 2048 1000 35.29
> 35.29 35.29
> 1 2048 1000 34.73
> 34.73 34.73
>
> 0 4096 1000 47.23
> 47.23 47.23
> 1 4096 1000 46.55
> 46.56 46.56
>
> 0 8192 1000 72.81
> 72.83 72.82
> 1 8192 1000 71.28
> 71.30 71.29
>
> 0 16384 1000 122.72
> 122.74 122.73
> 1 16384 1000 124.11
> 124.14 124.13
>
> 0 32768 1000 202.91
> 202.95 202.93
> 1 32768 1000 198.72
> 198.73 198.73
>
> 0 65536 640 326.64
> 326.69 326.66
> 1 65536 640 326.41
> 326.43 326.42
>
> 0 131072 320 578.43
> 578.54 578.49
> 1 131072 320 585.07
> 585.09 585.08
>
> 0 262144 160 1275.05
> 1275.28 1275.16
> 1 262144 160 1245.09
> 1245.16 1245.13
>
> 0 524288 80 2380.48
> 2380.94 2380.71
> 1 524288 80 2413.58
> 2413.71 2413.64
>
> 0 1048576 40 4665.67
> 4666.55 4666.11
> 1 1048576 40 4644.95
> 4645.05 4645.00
>
> 0 2097152 20 9234.45
> 9236.45 9235.45
> 1 2097152 20 9316.60
> 9317.00 9316.80
>
> 0 4194304 10 18342.80
> 18345.80 18344.30
> 1 4194304 10 18693.79
> 18694.31 18694.05
>
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.03
> 0.04 0.04
>
> 0 4 1000 30.58
> 30.61 30.59
>
> 0 8 1000 25.96
> 25.98 25.97
>
> 0 16 1000 25.81
> 25.82 25.82
>
> 0 32 1000 26.25
> 26.27 26.26
>
> 0 64 1000 30.18
> 30.20 30.19
>
> 0 128 1000 31.60
> 31.62 31.61
>
> 0 256 1000 35.04
> 35.06 35.05
>
> 0 512 1000 38.56
> 38.58 38.57
>
> 0 1024 1000 47.80
> 47.83 47.81
>
> 0 2048 1000 60.04
> 60.07 60.05
>
> 0 4096 1000 83.09
> 83.13 83.11
>
> 0 8192 1000 133.29
> 133.34 133.31
>
> 0 16384 1000 242.23
> 242.32 242.26
>
> 0 32768 1000 526.61
> 526.70 526.64
>
> 0 65536 640 882.81
> 882.97 882.85
>
> 0 131072 320 1624.91
> 1625.51 1625.14
>
> 0 262144 160 4753.44
> 4755.68 4754.38
>
> 0 524288 80 9000.99
> 9010.04 9005.16
>
> 0 1048576 40 17733.20
> 17769.92 17750.84
>
> 0 2097152 20 34719.90
> 34861.45 34789.24
>
> 0 4194304 10 68189.00
> 68761.81 68472.36
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Bcast
> Test Duration: 4.579 seconds
> Test Performances :
> #----------------------------------------------------------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.05 0.07 0.06
> 1 1000 14.75 14.76 14.75
> 2 1000 7.38 7.39 7.39
> 4 1000 7.33 7.35 7.34
> 8 1000 7.48 7.50 7.49
> 16 1000 7.45 7.46 7.46
> 32 1000 7.56 7.57 7.57
> 64 1000 8.76 8.78 8.77
> 128 1000 9.21 9.22 9.22
> 256 1000 9.94 9.95 9.94
> 512 1000 11.04 11.06 11.05
> 1024 1000 13.60 13.62 13.61
> 2048 1000 16.79 16.81 16.80
> 4096 1000 22.39 22.42 22.40
> 8192 1000 33.83 33.87 33.85
> 16384 1000 57.91 57.97 57.94
> 32768 1000 82.55 82.58 82.56
> 65536 640 126.93 126.95 126.94
> 131072 320 216.22 216.27 216.25
> 262144 160 391.68 391.83 391.75
> 524288 80 748.29 748.54 748.41
> 1048576 40 1450.13 1450.65 1450.39
> 2097152 20 2889.35 2890.60 2889.98
> 4194304 10 5725.70 5728.20 5726.95
>
> #----------------------------------------------------------------
> # ( #processes = 4 )
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.05 0.08 0.06
> 1 1000 102.54 102.57 102.55
> 2 1000 12.27 12.28 12.28
> 4 1000 12.01 12.03 12.02
> 8 1000 12.15 12.17 12.16
> 16 1000 12.14 12.15 12.14
> 32 1000 13.18 13.20 13.18
> 64 1000 13.45 13.47 13.46
> 128 1000 13.42 13.44 13.43
> 256 1000 14.15 14.16 14.15
> 512 1000 15.26 15.28 15.27
> 1024 1000 18.22 18.25 18.23
> 2048 1000 23.47 23.51 23.49
> 4096 1000 33.43 33.47 33.45
> 8192 1000 55.04 55.10 55.07
> 16384 1000 106.79 106.93 106.86
> 32768 1000 209.48 209.62 209.56
> 65536 640 345.42 345.71 345.58
> 131072 320 614.53 615.37 614.97
> 262144 160 1154.42 1157.07 1155.78
> 524288 80 2227.77 2237.69 2232.80
> 1048576 40 4380.35 4417.95 4399.30
> 2097152 20 8724.81 8872.44 8798.98
> 4194304 10 17405.30 17974.09 17690.14
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Bcast -multi 1
> Test Duration: 4.469 seconds
> Test Performances :
> #-------------------------------------------------------------
> ----------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.05
> 0.05 0.05
> 1 0 1000 0.05
> 0.06 0.06
>
> 0 1 1000 14.78
> 14.80 14.79
> 1 1 1000 14.88
> 14.88 14.88
>
> 0 2 1000 7.23
> 7.24 7.24
> 1 2 1000 7.11
> 7.12 7.12
>
> 0 4 1000 7.18
> 7.20 7.19
> 1 4 1000 7.13
> 7.13 7.13
>
> 0 8 1000 7.31
> 7.32 7.32
> 1 8 1000 7.27
> 7.27 7.27
>
> 0 16 1000 7.35
> 7.36 7.36
> 1 16 1000 7.31
> 7.31 7.31
>
> 0 32 1000 7.51
> 7.52 7.51
> 1 32 1000 7.45
> 7.45 7.45
>
> 0 64 1000 8.66
> 8.67 8.66
> 1 64 1000 8.65
> 8.66 8.65
>
> 0 128 1000 10.88
> 10.89 10.88
> 1 128 1000 8.95
> 8.95 8.95
>
> 0 256 1000 9.78
> 9.80 9.79
> 1 256 1000 9.60
> 9.60 9.60
>
> 0 512 1000 10.98
> 11.00 10.99
> 1 512 1000 10.84
> 10.84 10.84
>
> 0 1024 1000 13.53
> 13.55 13.54
> 1 1024 1000 13.35
> 13.36 13.36
>
> 0 2048 1000 16.76
> 16.80 16.78
> 1 2048 1000 16.90
> 16.91 16.90
>
> 0 4096 1000 22.17
> 22.20 22.19
> 1 4096 1000 22.82
> 22.84 22.83
>
> 0 8192 1000 35.69
> 35.73 35.71
> 1 8192 1000 33.15
> 33.19 33.17
>
> 0 16384 1000 56.78
> 56.84 56.81
> 1 16384 1000 56.69
> 56.73 56.71
>
> 0 32768 1000 82.71
> 82.74 82.73
> 1 32768 1000 81.19
> 81.21 81.20
>
> 0 65536 640 126.77
> 126.80 126.79
> 1 65536 640 125.70
> 125.70 125.70
>
> 0 131072 320 217.37
> 217.43 217.40
> 1 131072 320 215.39
> 215.44 215.42
>
> 0 262144 160 389.88
> 390.01 389.94
> 1 262144 160 393.20
> 393.21 393.21
>
> 0 524288 80 742.11
> 742.35 742.23
> 1 524288 80 742.81
> 742.82 742.82
>
> 0 1048576 40 1458.05
> 1458.50 1458.27
> 1 1048576 40 1454.42
> 1454.45 1454.44
>
> 0 2097152 20 2890.50
> 2891.86 2891.18
> 1 2097152 20 2855.85
> 2856.10 2855.97
>
> 0 4194304 10 5732.99
> 5735.29 5734.14
> 1 4194304 10 5688.50
> 5689.40 5688.95
>
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.05
> 0.08 0.06
>
> 0 1 1000 23.53
> 23.55 23.54
>
> 0 2 1000 12.26
> 12.28 12.27
>
> 0 4 1000 12.09
> 12.11 12.10
>
> 0 8 1000 12.25
> 12.27 12.26
>
> 0 16 1000 12.24
> 12.26 12.25
>
> 0 32 1000 12.54
> 12.56 12.55
>
> 0 64 1000 12.86
> 12.88 12.87
>
> 0 128 1000 13.22
> 13.24 13.23
>
> 0 256 1000 14.00
> 14.02 14.01
>
> 0 512 1000 15.28
> 15.30 15.29
>
> 0 1024 1000 18.15
> 18.18 18.16
>
> 0 2048 1000 23.20
> 23.24 23.21
>
> 0 4096 1000 33.01
> 33.05 33.03
>
> 0 8192 1000 54.58
> 54.64 54.61
>
> 0 16384 1000 104.19
> 104.32 104.25
>
> 0 32768 1000 205.32
> 205.45 205.39
>
> 0 65536 640 340.68
> 340.97 340.84
>
> 0 131072 320 608.96
> 609.78 609.38
>
> 0 262144 160 1141.46
> 1144.26 1142.90
>
> 0 524288 80 2199.45
> 2209.40 2204.53
>
> 0 1048576 40 4309.53
> 4346.97 4328.42
>
> 0 2097152 20 8464.45
> 8617.40 8541.36
>
> 0 4194304 10 16601.39
> 17169.90 16886.12
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Allgather
> Test Duration: 13.729 seconds
> Test Performances :
> #----------------------------------------------------------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.03 0.03 0.03
> 1 1000 15.19 15.31 15.25
> 2 1000 14.84 14.85 14.85
> 4 1000 14.87 14.87 14.87
> 8 1000 15.19 15.20 15.19
> 16 1000 15.05 15.05 15.05
> 32 1000 17.04 17.04 17.04
> 64 1000 17.77 17.78 17.78
> 128 1000 19.10 19.11 19.11
> 256 1000 20.90 20.90 20.90
> 512 1000 24.58 24.58 24.58
> 1024 1000 31.63 31.63 31.63
> 2048 1000 38.65 38.65 38.65
> 4096 1000 56.61 56.62 56.61
> 8192 1000 89.13 89.16 89.15
> 16384 1000 152.15 152.18 152.17
> 32768 1000 222.54 222.58 222.56
> 65536 640 382.26 382.32 382.29
> 131072 320 693.92 694.03 693.98
> 262144 160 1431.36 1431.64 1431.50
> 524288 80 2969.54 2970.19 2969.86
> 1048576 40 5996.80 5997.12 5996.96
> 2097152 20 12119.86 12122.35 12121.10
> 4194304 10 24130.31 24134.80 24132.55
>
> #----------------------------------------------------------------
> # ( #processes = 4 )
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.03 0.03 0.03
> 1 1000 28.14 28.16 28.15
> 2 1000 24.09 24.11 24.10
> 4 1000 24.15 24.17 24.16
> 8 1000 24.39 24.42 24.40
> 16 1000 26.29 26.31 26.30
> 32 1000 27.47 27.49 27.48
> 64 1000 29.99 30.01 30.00
> 128 1000 32.99 33.01 32.99
> 256 1000 38.72 38.75 38.74
> 512 1000 46.71 46.74 46.72
> 1024 1000 60.89 60.93 60.91
> 2048 1000 91.27 91.32 91.29
> 4096 1000 156.02 156.11 156.06
> 8192 1000 267.91 267.98 267.94
> 16384 1000 458.11 458.20 458.14
> 32768 1000 842.71 842.89 842.77
> 65536 640 1525.32 1525.87 1525.54
> 131072 320 2875.53 2877.73 2876.53
> 262144 160 5612.53 5621.31 5616.71
> 524288 80 10901.43 10936.58 10918.59
> 1048576 40 21652.57 21793.70 21722.27
> 2097152 20 42803.20 43370.85 43085.33
> 4194304 10 84603.71 86869.99 85733.28
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Allgather -multi 1
> Test Duration: 13.619 seconds
> Test Performances :
> #-------------------------------------------------------------
> ----------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.03
> 0.03 0.03
> 1 0 1000 0.03
> 0.03 0.03
>
> 0 1 1000 14.79
> 15.04 14.92
> 1 1 1000 23.05
> 23.06 23.06
>
> 0 2 1000 14.37
> 14.37 14.37
> 1 2 1000 15.05
> 15.06 15.05
>
> 0 4 1000 14.51
> 14.51 14.51
> 1 4 1000 15.14
> 15.15 15.14
>
> 0 8 1000 14.66
> 14.66 14.66
> 1 8 1000 15.31
> 15.32 15.32
>
> 0 16 1000 14.82
> 14.82 14.82
> 1 16 1000 15.51
> 15.53 15.52
>
> 0 32 1000 16.16
> 16.16 16.16
> 1 32 1000 16.74
> 16.75 16.74
>
> 0 64 1000 17.67
> 17.67 17.67
> 1 64 1000 18.40
> 18.41 18.40
>
> 0 128 1000 20.54
> 20.54 20.54
> 1 128 1000 19.50
> 19.50 19.50
>
> 0 256 1000 20.78
> 20.78 20.78
> 1 256 1000 21.49
> 21.50 21.50
>
> 0 512 1000 24.57
> 24.57 24.57
> 1 512 1000 24.92
> 24.93 24.92
>
> 0 1024 1000 30.24
> 30.25 30.25
> 1 1024 1000 30.48
> 30.50 30.49
>
> 0 2048 1000 38.88
> 38.88 38.88
> 1 2048 1000 39.24
> 39.27 39.25
>
> 0 4096 1000 54.90
> 54.91 54.90
> 1 4096 1000 56.42
> 56.45 56.44
>
> 0 8192 1000 92.50
> 92.53 92.51
> 1 8192 1000 87.64
> 87.70 87.67
>
> 0 16384 1000 151.29
> 151.33 151.31
> 1 16384 1000 150.74
> 150.75 150.74
>
> 0 32768 1000 223.62
> 223.66 223.64
> 1 32768 1000 224.75
> 224.76 224.75
>
> 0 65536 640 375.88
> 375.95 375.92
> 1 65536 640 380.39
> 380.40 380.39
>
> 0 131072 320 691.63
> 691.77 691.70
> 1 131072 320 698.52
> 698.55 698.53
>
> 0 262144 160 1428.96
> 1429.24 1429.10
> 1 262144 160 1455.81
> 1455.85 1455.83
>
> 0 524288 80 3038.20
> 3038.79 3038.49
> 1 524288 80 3035.49
> 3035.53 3035.51
>
> 0 1048576 40 6025.07
> 6026.22 6025.65
> 1 1048576 40 6220.52
> 6220.67 6220.60
>
> 0 2097152 20 12213.95
> 12216.20 12215.07
> 1 2097152 20 12457.41
> 12457.70 12457.55
>
> 0 4194304 10 24338.40
> 24342.31 24340.35
> 1 4194304 10 24826.29
> 24826.71 24826.50
>
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.03
> 0.04 0.03
>
> 0 1 1000 27.80
> 27.82 27.80
>
> 0 2 1000 24.05
> 24.07 24.06
>
> 0 4 1000 23.99
> 24.01 24.00
>
> 0 8 1000 24.73
> 24.76 24.74
>
> 0 16 1000 26.14
> 26.17 26.15
>
> 0 32 1000 27.19
> 27.21 27.20
>
> 0 64 1000 29.54
> 29.56 29.55
>
> 0 128 1000 32.49
> 32.51 32.50
>
> 0 256 1000 38.65
> 38.67 38.66
>
> 0 512 1000 47.00
> 47.03 47.02
>
> 0 1024 1000 61.25
> 61.29 61.27
>
> 0 2048 1000 92.54
> 92.59 92.57
>
> 0 4096 1000 155.10
> 155.19 155.14
>
> 0 8192 1000 264.47
> 264.54 264.49
>
> 0 16384 1000 450.72
> 450.82 450.75
>
> 0 32768 1000 844.25
> 844.43 844.31
>
> 0 65536 640 1521.38
> 1521.93 1521.59
>
> 0 131072 320 2870.66
> 2872.85 2871.64
>
> 0 262144 160 5624.76
> 5633.56 5628.95
>
> 0 524288 80 10923.28
> 10958.54 10940.53
>
> 0 1048576 40 21651.00
> 21792.45 21720.84
>
> 0 2097152 20 42745.60
> 43313.50 43027.81
>
> 0 4194304 10 84495.00
> 86766.01 85626.75
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Allgatherrv
> Test Duration: 1.080 seconds
> Test Performances : No output for this performance test.
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Allgatherrv -multi 1
> Test Duration: 1.359 seconds
> Test Performances : No output for this performance test.
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-W: Could not run
> /data/home/scott/builds/TopspinOS-2.6.0/build183/protest/Igr7/
> 050906_002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB-MPI1
> Alltoall : 0
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Alltoall
> Test Duration: 64.459 seconds
> Test Performances :
> #----------------------------------------------------------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.03 0.03 0.03
> 1 1000 8.12 8.18 8.15
> 2 1000 7.80 7.80 7.80
> 4 1000 7.74 7.75 7.74
> 8 1000 7.87 7.88 7.88
> 16 1000 8.04 8.05 8.04
> 32 1000 7.96 7.96 7.96
> 64 1000 9.30 9.31 9.30
> 128 1000 10.37 10.38 10.38
> 256 1000 12.03 12.04 12.04
> 512 1000 12.83 12.85 12.84
> 1024 1000 16.23 16.25 16.24
> 2048 1000 20.55 20.57 20.56
> 4096 1000 26.89 26.89 26.89
> 8192 1000 44.50 44.51 44.51
> 16384 1000 78.13 78.18 78.15
> 32768 1000 132.28 132.31 132.30
> 65536 640 222.20 222.25 222.23
> 131072 320 405.07 405.15 405.11
> 262144 160 1069.56 1069.62 1069.59
> 524288 80 2479.89 2481.27 2480.58
> 1048576 40 4875.95 4882.20 4879.08
> 2097152 20 9520.25 9530.10 9525.18
> 4194304 10 19020.81 19048.00 19034.41
>
> #----------------------------------------------------------------
> # ( #processes = 4 )
> #----------------------------------------------------------------
> #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 0 1000 0.03 0.03 0.03
> 1 1000 27.06 27.07 27.07
> 2 1000 16.48 16.49 16.48
> 4 1000 17.28 17.29 17.28
> 8 1000 20.79 20.80 20.80
> 16 1000 18.60 18.61 18.61
> 32 1000 18.46 18.47 18.47
> 64 1000 19.09 19.10 19.09
> 128 1000 16.24 16.29 16.26
> 256 1000 18.10 18.11 18.10
> 512 1000 19.39 19.40 19.39
> 1024 1000 26.51 26.52 26.52
> 2048 1000 37.85 37.86 37.85
> 4096 1000 60.41 60.42 60.41
> 8192 1000 103.04 103.08 103.06
> 16384 1000 200.75 200.79 200.78
> 32768 1000 2981.05 2982.04 2981.70
> 65536 640 3435.51 3436.92 3436.32
> 131072 320 4224.32 4226.30 4225.34
> 262144 160 6300.70 6305.17 6303.13
> Signal:11 info.si_errno:0(Success) si_code:1(SEGV_MAPERR)
> Failing at addr:0x20
> [0]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel4-i686-mvapi/lib/
> libopal.so.0 [0xdfd66c]
> [1] func:/lib/tls/libpthread.so.0 [0x938e40]
> [2]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_rcache_rb.so(mca_rcache_rb_insert+0x30) [0x85012c]
> [3]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_mpool_mvapi.so(mca_mpool_mvapi_register+0x1b3) [0x9d3157]
> [4]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so(mca_pml_ob1_rdma_btls+0x18c) [0x292458]
> [5]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so [0x2951f9]
> [6]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so(mca_pml_ob1_recv_request_progress+0x5e4
> ) [0x294098]
> [7]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so(mca_pml_ob1_recv_request_match_specific
> +0xb0) [0x2948f0]
> [8]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so(mca_pml_ob1_irecv+0x139) [0x290d01]
> [9]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_coll_tuned.so(ompi_coll_tuned_sendrecv_actual+0x38)
[0x8a0394]
> [10]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_coll_tuned.so(ompi_coll_tuned_alltoall_intra_pairwise+0xb4)
[0x8a1550]
> [11]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_coll_tuned.so(ompi_coll_tuned_alltoall_intra_dec_fixed+0x7d)
[0x8a0531]
> [12]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel4-i686-mvapi/lib/
> libmpi.so.0(MPI_Alltoall+0x167) [0xf89daf]
> [13]
> func:/data/home/scott/builds/TopspinOS-2.6.0/build183/protest/
> Igr7/050906_002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB
> -MPI1(Alltoall+0xa4) [0x804dd28]
> [14]
> func:/data/home/scott/builds/TopspinOS-2.6.0/build183/protest/
> Igr7/050906_002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB
> -MPI1(main+0x32e) [0x804a78a]
> [15] func:/lib/tls/libc.so.6(__libc_start_main+0xf8) [0x126748]
> [16]
> func:/data/home/scott/builds/TopspinOS-2.6.0/build183/protest/
> Igr7/050906_002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB
> -MPI1(free+0x41) [0x804a3cd]
> *** End of error message ***
> Broken pipe
> _releng_at_svbu-qa2650-16:/data/home/scott/builds/TopspinOS-2.6.
> 0/build183/protest/Igr7/050906_002959/open.pallas/1147159823/P
> MB2.2.1/SRC_PMB\[releng_at_svbu-qa2650-16 SRC_PMB]$
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-W: Could not run
> /data/home/scott/builds/TopspinOS-2.6.0/build183/protest/Igr7/
> 050906_002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB-MPI1
> Alltoall -multi 1: 0
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Alltoall -multi 1
> Test Duration: 64.529 seconds
> Test Performances :
> #-------------------------------------------------------------
> ----------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.03
> 0.03 0.03
> 1 0 1000 0.03
> 0.03 0.03
>
> 0 1 1000 8.08
> 9.82 8.95
> 1 1 1000 16.16
> 16.16 16.16
>
> 0 2 1000 7.75
> 7.75 7.75
> 1 2 1000 7.72
> 7.74 7.73
>
> 0 4 1000 7.69
> 7.69 7.69
> 1 4 1000 7.70
> 7.71 7.71
>
> 0 8 1000 7.83
> 7.83 7.83
> 1 8 1000 8.58
> 8.58 8.58
>
> 0 16 1000 7.88
> 7.88 7.88
> 1 16 1000 7.95
> 7.95 7.95
>
> 0 32 1000 8.19
> 8.20 8.20
> 1 32 1000 8.05
> 8.05 8.05
>
> 0 64 1000 9.34
> 9.34 9.34
> 1 64 1000 9.29
> 9.30 9.30
>
> 0 128 1000 11.69
> 11.69 11.69
> 1 128 1000 9.74
> 9.75 9.75
>
> 0 256 1000 11.16
> 11.16 11.16
> 1 256 1000 10.95
> 10.96 10.95
>
> 0 512 1000 12.38
> 12.39 12.39
> 1 512 1000 12.10
> 12.10 12.10
>
> 0 1024 1000 16.07
> 16.08 16.07
> 1 1024 1000 15.47
> 15.48 15.48
>
> 0 2048 1000 20.24
> 20.26 20.25
> 1 2048 1000 20.30
> 20.32 20.31
>
> 0 4096 1000 28.03
> 28.04 28.04
> 1 4096 1000 27.56
> 27.56 27.56
>
> 0 8192 1000 45.71
> 45.72 45.72
> 1 8192 1000 41.91
> 41.92 41.92
>
> 0 16384 1000 79.02
> 79.03 79.02
> 1 16384 1000 77.04
> 77.07 77.05
>
> 0 32768 1000 131.96
> 132.00 131.98
> 1 32768 1000 128.89
> 128.89 128.89
>
> 0 65536 640 224.41
> 224.46 224.43
> 1 65536 640 219.07
> 219.08 219.07
>
> 0 131072 320 517.59
> 518.01 517.80
> 1 131072 320 402.72
> 402.77 402.75
>
> 0 262144 160 1071.88
> 1072.39 1072.14
> 1 262144 160 1041.22
> 1041.33 1041.27
>
> 0 524288 80 2321.41
> 2321.47 2321.44
> 1 524288 80 2391.01
> 2391.20 2391.11
>
> 0 1048576 40 4720.67
> 4722.40 4721.54
> 1 1048576 40 4707.35
> 4707.55 4707.45
>
> 0 2097152 20 9546.55
> 9559.10 9552.82
> 1 2097152 20 9685.34
> 9693.84 9689.59
>
> 0 4194304 10 19362.09
> 19431.60 19396.84
> 1 4194304 10 19253.30
> 19280.50 19266.90
>
> #-------------------------------------------------------------
> ----------------
> # ( #processes = 4 )
> #-------------------------------------------------------------
> ----------------
> Group #bytes #repetitions t_min[usec]
> t_max[usec] t_avg[usec]
>
> 0 0 1000 0.03
> 0.03 0.03
>
> 0 1 1000 24.88
> 24.89 24.89
>
> 0 2 1000 16.25
> 16.27 16.26
>
> 0 4 1000 17.32
> 17.33 17.33
>
> 0 8 1000 17.01
> 17.02 17.01
>
> 0 16 1000 17.76
> 17.77 17.76
>
> 0 32 1000 18.88
> 18.88 18.88
>
> 0 64 1000 22.38
> 22.39 22.39
>
> 0 128 1000 15.94
> 15.95 15.95
>
> 0 256 1000 17.82
> 17.84 17.83
>
> 0 512 1000 19.05
> 19.06 19.05
>
> 0 1024 1000 26.06
> 26.08 26.07
>
> 0 2048 1000 37.71
> 37.73 37.72
>
> 0 4096 1000 59.67
> 59.69 59.68
>
> 0 8192 1000 105.69
> 105.72 105.70
>
> 0 16384 1000 210.82
> 210.87 210.84
>
> 0 32768 1000 3002.53
> 3003.31 3002.91
>
> 0 65536 640 3572.82
> 3574.32 3573.68
>
> 0 131072 320 4486.19
> 4488.97 4487.67
>
> 0 262144 160 6401.08
> 6408.27 6404.88
> Signal:11 info.si_errno:0(Success) si_code:1(SEGV_MAPERR)
> Failing at addr:0x20
> [0]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel4-i686-mvapi/lib/
> libopal.so.0 [0x2f566c]
> [1] func:/lib/tls/libpthread.so.0 [0x65ee40]
> [2]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_rcache_rb.so(mca_rcache_rb_insert+0x30) [0xcc112c]
> [3]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_mpool_mvapi.so(mca_mpool_mvapi_register+0x1b3) [0x266157]
> [4]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so(mca_pml_ob1_rdma_btls+0x18c) [0x3d7458]
> [5]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so [0x3da1f9]
> [6]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so(mca_pml_ob1_recv_request_progress+0x5e4
> ) [0x3d9098]
> [7]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so(mca_pml_ob1_recv_request_match_specific
> +0xb0) [0x3d98f0]
> [8]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
> openmpi/mca_pml_ob1.so(mca_pml_ob1_irecv+0x139) [0x3d5d01]
> [9]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_coll_tuned.so(ompi_coll_tuned_sendrecv_actual+0x38)
[0x2a5394]
> [10]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_coll_tuned.so(ompi_coll_tuned_alltoall_intra_pairwise+0xb4)
[0x2a6550]
> [11]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel3-i686-mvapi/lib/
openmpi/mca_coll_tuned.so(ompi_coll_tuned_alltoall_intra_dec_fixed+0x7d)
[0x2a5531]
> [12]
> func:/data/software/qa/MPI/openmpi-1.1a2-rhel4-i686-mvapi/lib/
> libmpi.so.0(MPI_Alltoall+0x167) [0x492daf]
> [13]
> func:/data/home/scott/builds/TopspinOS-2.6.0/build183/protest/
> Igr7/050906_002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB
> -MPI1(Alltoall+0xa4) [0x804dd28]
> [14]
> func:/data/home/scott/builds/TopspinOS-2.6.0/build183/protest/
> Igr7/050906_002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB
> -MPI1(main+0x32e) [0x804a78a]
> [15] func:/lib/tls/libc.so.6(__libc_start_main+0xf8) [0x129748]
> [16]
> func:/data/home/scott/builds/TopspinOS-2.6.0/build183/protest/
> Igr7/050906_002959/open.pallas/1147159823/PMB2.2.1/SRC_PMB/PMB
> -MPI1(free+0x41) [0x804a3cd]
> *** End of error message ***
> Broken pipe
> _releng_at_svbu-qa2650-16:/data/home/scott/builds/TopspinOS-2.6.
> 0/build183/protest/Igr7/050906_002959/open.pallas/1147159823/P
> MB2.2.1/SRC_PMB\[releng_at_svbu-qa2650-16 SRC_PMB]$
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Barrier
> Test Duration: 1.440 seconds
> Test Performances :
> #---------------------------------------------------
> # ( #processes = 2 )
> # ( 2 additional processes waiting in MPI_Barrier)
> #---------------------------------------------------
> #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 1000 19.04 19.05 19.05
>
> #---------------------------------------------------
> # ( #processes = 4 )
> #---------------------------------------------------
> #repetitions t_min[usec] t_max[usec] t_avg[usec]
> 1000 40.79 40.81 40.80
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> ###TEST-D: Open-MPI VAPI Test on hosts svbu-qa2650-16
> svbu-qa2650-15 svbu-qa2650-14 svbu-qa2650-13
> Number of hosts : 4
> Number of processes : 4
> Test Type: Pallas MPI Benchmark / Barrier -multi 1
> Test Duration: 1.240 seconds
> Test Performances :
> #----------------------------------------------------------------
> # ( 2 groups of 2 processes each running simultaneous )
> # Group 0 : 0 1
> # Group 1 : 2 3
> #----------------------------------------------------------------
> Group #repetitions t_min[usec] t_max[usec] t_avg[usec]
>
> 0 1000 19.02 19.02 19.02
> 1 1000 17.82 17.83 17.83
>
> #----------------------------------------------------------------
> # ( #processes = 4 )
> #----------------------------------------------------------------
> Group #repetitions t_min[usec] t_max[usec] t_avg[usec]
>
> 0 1000 40.64 40.66 40.65
> svbu-qa2650-16 CPU Usage: 35.10%
> svbu-qa2650-15 CPU Usage: 32.12%
> svbu-qa2650-14 CPU Usage: 31.84%
> svbu-qa2650-13 CPU Usage: 31.70%
>
> InfiniBand.Performance.MPI.Pallas.Open-MPI.VAPI.GnuC --> FAIL
>
> End of protest run
>