STREAM "Experimental & Nonstandard" results

This is the "Experimental & Nonstandard" set of STREAM benchmark results.

This set includes only the following categories:

Assembly language results.
Other hand-coded results.
Simulated results.
Results from experimental or non-production hardware.
Results from "partially depopulated" systems (i.e., results based on the use of a subset of a system's processors, where the processors are "spread around" to decrease contention).

STREAM Memory MFLOPS --- John D. McCalpin, mccalpin@cs.virginia.edu
Revised to Tue Apr 15 17:23:56 CDT 2008

All results are in MFLOPS --- 1 MFLOPS=10^6 FP OPS/sec

----------------------------------------------------------------
Machine ID                      ncpus    SCALE      ADD    TRIAD
----------------------------------------------------------------
Azul_Vega2_7280                  768   5201.5   3781.5   7616.2 data
Cray_J98                           4    297.1    182.8    362.2
Cray_J98                           2    163.1    102.6    202.8
Cray_J98                           1     85.4     52.5    101.0
Cray_J98                           1     87.2     52.6    106.2
IBM_eServer_p680                  24    284.8    208.7    419.7 data
IBM_eServer_p680                  24    427.8    281.0    566.4 data
IBM_eServer_p680                  22    282.2    206.3    417.0 data
IBM_eServer_p680                  20    268.6    190.9    384.4 data
IBM_eServer_p680                  18    257.6    185.4    373.6 data
IBM_eServer_p680                  16    235.8    163.9    329.5 data
IBM_eServer_p680                  14    214.9    152.0    305.8 data
IBM_eServer_p680                  12    181.9    124.5    250.3 data
IBM_eServer_p680                  10    165.3    115.7    232.9 data
IBM_eServer_p680                   8    127.3     87.5    176.0 data
IBM_eServer_p680                   6    102.1     71.8    144.5 data
IBM_eServer_p680                   4     72.6     48.5     97.1 data
IBM_eServer_p680                   2     37.7     26.0     52.1 data
IBM_eServer_p680                   1     19.0     13.1     26.1 data
IBM_eServer_p680                   1     26.5     16.8     33.6 data
SGI_Origin2000-195_1_per_node     32    541.9    405.5    786.2
SGI_Origin2000-195_1_per_node     16    291.1    192.9    405.9
SGI_Origin2000-195_1_per_node      8    148.2     99.3    195.6
SGI_Origin2000-195_1_per_node      4     73.8     51.7    106.9
SGI_Origin2000-250_1_per_node     32    668.8    475.0    908.3
SGI_Origin2000-250_1_per_node     16    340.6    236.2    436.7
SGI_Origin2000-250_1_per_node      8    168.8    118.8    228.3
SGI_Origin2000-250_1_per_node      4     84.4     58.8    125.8
SGI_Origin2000-250_1_per_node      2     42.5     28.7     59.5
Sun_UE_10000_400_1_per_node       16    218.3    149.2    232.4
Sun_UE_10000_400_2_per_node       32    368.9    254.0    431.7
Sun_UE_10000_400_2_per_node       16    180.3    144.4    287.2
Sun_UE_10000_400_3_per_node       48    462.2    326.0    602.5
TMC_CM-200_64k_10               2048   2625.7   1977.1   3768.8
TMC_CM-200__8k_10_MHz            256    328.2    247.1    471.1
TMC_CM-2_64k_7_MHz              2048   1612.2   1219.1   2329.9
TMC_CM-2_64k_8_MHz              2048   1912.0   1396.7   2666.7
TMC_CM-2__8k_7_MHz               256    201.5    152.4    291.2
TMC_CM-2__8k_8_MHz               256    239.0    174.6    333.3
Tera_MTA_simulated                16   1877.1   1273.0   2542.5
Tera_MTA_simulated                 8    937.3    633.0   1270.1
Tera_MTA_simulated                 4    480.0    324.9    650.0
Tera_MTA_simulated                 2    250.0    169.1    338.5
Tera_MTA_simulated                 1    128.8     87.1    174.3
----------------------------------------------------------