SHOW:
|
|
- or go back to the newest paste.
1 | [0] DAPL startup(): trying to open default DAPL provider from dat registry: ibnic0v2 | |
2 | [1] DAPL startup(): trying to open default DAPL provider from dat registry: ibnic0v2 | |
3 | [0] MPI startup(): DAPL provider ibnic0v2 | |
4 | [1] MPI startup(): DAPL provider ibnic0v2 | |
5 | [0] MPI startup(): dapl data transfer mode | |
6 | [1] MPI startup(): dapl data transfer mode | |
7 | [0] MPI startup(): Internal info: pinning initialization was done | |
8 | [0] MPI startup(): Rank Pid Node name Pin cpu | |
9 | ||
10 | [0] MPI startup(): 0 3700 CN01 {0,1,2,3,4,5,6,7} | |
11 | ||
12 | [0] MPI startup(): 1 2632 CN02 {0,1,2,3,4,5,6,7} | |
13 | ||
14 | [1] MPI startup(): Internal info: pinning initialization was done | |
15 | [0] MPI startup(): I_MPI_DEBUG=5 | |
16 | [0] MPI startup(): I_MPI_PIN_MAPPING=1:0 0 | |
17 | #--------------------------------------------------- | |
18 | # Intel (R) MPI Benchmark Suite V3.2.3, MPI-1 part | |
19 | #--------------------------------------------------- | |
20 | # Date : Thu Jan 17 08:56:05 2013 | |
21 | # Machine : Intel(R) 64 Family 6 Model 26 Stepping 5, GenuineIntel | |
22 | # Release : 6.1.7601 | |
23 | # Version : Service Pack 1 | |
24 | # MPI Version : 2.2 | |
25 | # MPI Thread Environment: MPI_THREAD_MULTIPLE | |
26 | ||
27 | ||
28 | # New default behavior from Version 3.2 on: | |
29 | ||
30 | # the number of iterations per message size is cut down | |
31 | # dynamically when a certain run time (per message size sample) | |
32 | # is expected to be exceeded. Time limit is defined by variable | |
33 | # "SECS_PER_SAMPLE" (=> IMB_settings.h) | |
34 | # or through the flag => -time | |
35 | ||
36 | ||
37 | ||
38 | # Calling sequence was: | |
39 | ||
40 | # C:\Users\sg\Desktop\imb_3.2.3\WINDOWS\IMB-MPI1_VS_2010\x64\Release\IMB-MPI1.exe | |
41 | ||
42 | # Minimum message length in bytes: 0 | |
43 | # Maximum message length in bytes: 4194304 | |
44 | # | |
45 | # MPI_Datatype : MPI_BYTE | |
46 | # MPI_Datatype for reductions : MPI_FLOAT | |
47 | # MPI_Op : MPI_SUM | |
48 | # | |
49 | # | |
50 | ||
51 | # List of Benchmarks to run: | |
52 | ||
53 | # PingPong | |
54 | # PingPing | |
55 | # Sendrecv | |
56 | # Exchange | |
57 | # Allreduce | |
58 | # Reduce | |
59 | # Reduce_scatter | |
60 | # Allgather | |
61 | # Allgatherv | |
62 | # Gather | |
63 | # Gatherv | |
64 | # Scatter | |
65 | # Scatterv | |
66 | # Alltoall | |
67 | # Alltoallv | |
68 | # Bcast | |
69 | # Barrier | |
70 | ||
71 | #--------------------------------------------------- | |
72 | # Benchmarking PingPong | |
73 | # #processes = 2 | |
74 | #--------------------------------------------------- | |
75 | #bytes #repetitions t[usec] Mbytes/sec | |
76 | 0 1000 3.99 0.00 | |
77 | 1 1000 3.99 0.24 | |
78 | 2 1000 3.76 0.51 | |
79 | 4 1000 3.77 1.01 | |
80 | 8 1000 3.78 2.02 | |
81 | 16 1000 3.81 4.01 | |
82 | 32 1000 3.93 7.77 | |
83 | 64 1000 3.93 15.52 | |
84 | 128 1000 4.05 30.12 | |
85 | 256 1000 4.10 59.57 | |
86 | 512 1000 4.41 110.62 | |
87 | 1024 1000 4.99 195.63 | |
88 | 2048 1000 6.22 314.13 | |
89 | 4096 1000 8.30 470.55 | |
90 | 8192 1000 10.63 735.28 | |
91 | 16384 1000 15.31 1020.76 | |
92 | 32768 1000 21.21 1473.49 | |
93 | 65536 640 31.53 1982.50 | |
94 | 131072 320 52.39 2385.87 | |
95 | 262144 160 94.76 2638.22 | |
96 | 524288 80 185.22 2699.49 | |
97 | 1048576 40 356.92 2801.73 | |
98 | 2097152 20 699.45 2859.38 | |
99 | 4194304 10 1393.73 2870.00 | |
100 | ||
101 | ||
102 | ... | |
103 | ||
104 | ||
105 | #---------------------------------------------------------------- | |
106 | # Benchmarking Bcast | |
107 | # #processes = 2 | |
108 | #---------------------------------------------------------------- | |
109 | #bytes #repetitions t_min[usec] t_max[usec] t_avg[usec] | |
110 | 0 1000 0.05 0.46 0.26 | |
111 | 1 1000 3.41 3.41 3.41 | |
112 | 2 1000 3.42 3.42 3.42 | |
113 | 4 1000 3.42 3.42 3.42 | |
114 | 8 1000 3.42 3.42 3.42 | |
115 | 16 1000 3.43 3.44 3.43 | |
116 | 32 1000 3.50 3.51 3.50 | |
117 | 64 1000 3.55 3.55 3.55 | |
118 | 128 1000 3.57 3.58 3.57 | |
119 | 256 1000 3.79 3.80 3.80 | |
120 | 512 1000 4.08 4.08 4.08 | |
121 | 1024 1000 4.74 4.75 4.74 | |
122 | 2048 1000 5.89 5.90 5.89 | |
123 | 4096 1000 8.12 8.13 8.13 | |
124 | 8192 1000 10.31 10.32 10.31 | |
125 | 16384 1000 14.74 14.75 14.74 | |
126 | 32768 1000 20.05 20.05 20.05 | |
127 | ||
128 | Fatal error in PMPI_Bcast: Other MPI error, error stack: | |
129 | PMPI_Bcast(2112)........: MPI_Bcast(buf=00000000030C0040, count=65536, MPI_BYTE, root=0, comm=0x84000000) failed | |
130 | MPIR_Bcast_impl(1670)...: | |
131 | I_MPIR_Bcast_intra(1887): Failure during collective | |
132 | MPIR_Bcast_intra(1461)..: | |
133 | MPIR_Bcast_binomial(156): message sizes do not match across processes in the collective | |
134 | - | [0:CN01.cual.local] unexpected disconnect completion event from [1:CN02.****.*****] |
134 | + | [0:CN01.****.*****] unexpected disconnect completion event from [1:CN02.****.*****] |
135 | Assertion failed in file .\dapl_conn_rc.c at line 1128: 0 | |
136 | internal ABORT - process 0 | |
137 | ||
138 | job aborted: | |
139 | rank: node: exit code[: error message] | |
140 | - | 0: CN01.cual.local: 1: process 0 exited without calling finalize |
140 | + | 0: CN01.****.*****: 1: process 0 exited without calling finalize |
141 | 1: CN02: 1: process 1 exited without calling finalize |