This set includes only 32-bit results --- i.e. results based on either 32-bit data elements or 32-bit arithmetic. Many of the 64-bit results actually move data in 32-bit chunks, but as long as the operands and operations are 64-bit, those cases remain in the other tables.
STREAM Memory MFLOPS --- John D. McCalpin, mccalpin@cs.virginia.edu Revised to Sun Oct 18 18:42:46 CDT 2009 All results are in MFLOPS --- 1 MFLOPS=10^6 FP OPS/sec ---------------------------------------------------------------- Machine ID ncpus SCALE ADD TRIAD ---------------------------------------------------------------- Alliant_VFX80 3 1.5 1.1 2.1 data Convex_C1XP 1 3.9 2.7 5.4 data Convex_C3210 1 5.8 4.5 9.1 Convex_C3220 2 11.1 8.8 17.4 Convex_C3410 1 9.4 6.5 12.9 Convex_C3420 2 14.9 11.0 21.7 DEC_4000-710 1 5.0 3.3 6.6 DEC_5000-200 1 1.3 0.9 1.5 data DEC_VAX_9000-420 1 5.1 3.3 6.5 FPS_510_EA_scalar 1 0.7 0.5 1.0 FPS_511_EA_vector 1 5.2 3.9 7.8 FPS_511_SPARC_scalar 1 1.9 1.1 2.2 FPS_511_SPARC_vector 1 5.3 4.0 7.9 FPS_MCP104 1 6.5 4.8 7.8 FPS_MCP728 7 29.9 16.4 32.9 HP_9000-720 1 2.9 2.0 4.3 HP_9000-730 1 2.3 1.9 3.3 data IBM_3090J-VF 1 10.4 7.2 14.5 IBM_3090J_scalar 1 3.5 3.0 5.7 IBM_RS-6000-250 1 3.2 2.2 4.4 IBM_RS6000-320 1 2.1 1.7 3.3 IBM_RS6000-320H 1 2.5 2.2 4.5 data IBM_RS6000-355 1 5.3 3.6 7.1 IBM_RS6000-530 1 2.8 2.6 5.0 IBM_RS6000-580 1 8.9 7.8 13.9 IBM_RS6000-950 1 4.7 4.3 8.3 NEC_SX-3-12 1 239.7 208.1 416.2 data NeXTStation_68040 1 0.5 0.4 0.7 Omron_Luna88k 1 0.8 0.7 1.1 data SGI_4D-240 1 0.9 0.8 1.3 SGI_4D-25 1 0.6 0.4 0.8 SGI_4D-35 1 1.8 1.6 2.1 SGI_Challenge_100 1 2.9 2.1 4.0 SGI_Challenge_150 1 4.2 2.8 5.3 SGI_Crimson 1 3.3 2.3 4.4 SGI_Indigo 1 1.3 1.1 2.2 Sun_4-490 1 0.9 1.0 1.3 Sun_670 1 1.3 0.9 2.0 data Sun_SS1- 1 0.5 0.5 0.7 Sun_SS10-30 1 2.6 1.6 2.8 Sun_SS10-41 1 2.4 1.5 3.2 Sun_SS2_(4-75) 1 1.1 0.8 1.7 data TMC_CM-5E 32 440.9 335.2 669.7 ----------------------------------------------------------