Memory Bandwidth STREAM Benchmark John D. McCalpin john@mccalpin.com Revised to: Wed Feb 8 14:23:59 1995 Bytes Bandwidth (MB/s) Machine /word Copy Scale Sum Triad --------------- ----- -------- -------- -------- -------- Cray YMP/C90 16 cpu 8 105497.4 104656.4 101736.1 103812.8 Cray YMP/C90 8 cpu 8 55071.9 55391.8 60843.3 63229.6 Cray YMP/C90 4 cpu 8 27610.3 27789.6 34633.3 35044.1 Cray YMP/C90 2 cpu 8 13866.0 13905.5 18233.2 18246.3 Cray YMP/C90 1 cpu 8 6965.4 6965.4 9378.7 9500.7 Cray Y/MP 8 cpu 8 19291.6 19294.2 26588.9 26802.2 Cray Y/MP 4 cpu 8 9685.8 9678.9 13781.4 13851.2 Cray Y/MP 8 2426.4 2426.2 3454.4 3396.9 Cray J94 4 cpu 8 4699.0 4754.0 4386.0 4346.0 Cray J94 2 cpu 8 2729.0 2609.0 2463.0 2433.0 Cray J94 1 cpu 8 1436.0 1395.0 1263.0 1275.0 Cray J94 1 cpu 8 1417.0 1366.0 1259.0 1212.0 Cray EL-98 8 cpu 8 2362.8 2310.5 2373.7 2363.8 Cray EL-98 4 cpu 8 1564.9 1569.8 1933.8 1955.5 Cray EL-98 2 cpu 8 826.7 833.8 1049.0 1078.2 Cray EL-98 1 cpu 8 437.2 436.7 536.2 476.8 Cray T3D 256 PEs assembly 8 98303.7 84229.5 57622.6 56078.7 Cray T3D 128 PEs assembly 8 49132.9 42113.9 28811.1 28032.5 Cray T3D 64 PEs assembly 8 24577.9 21061.6 14405.8 14020.2 Cray T3D 32 PEs assembly 8 12288.6 10530.7 7204.2 7010.7 Cray T3D 1 PEs assembly 8 384.5 329.4 225.4 220.1 Cray T3D 256 PEs Fortran 8 98316.2 84241.5 47824.1 45248.1 Cray T3D 128 PEs Fortran 8 49156.4 42128.2 23912.0 22625.2 Cray T3D 64 PEs Fortran 8 24580.7 21064.7 11955.7 11312.9 Cray T3D 32 PEs Fortran 8 12290.3 10533.4 5978.3 5656.4 Cray T3D 1 PEs Fortran 8 384.2 329.2 187.0 176.8 Cray CS6400, 32 cpu 8 824.1 819.6 885.0 882.6 Cray CS6400, 24 cpu 8 761.9 753.7 775.5 774.5 Cray CS6400, 16 cpu 8 611.5 601.0 596.0 594.6 Cray CS6400, 8 cpu 8 347.9 343.4 341.4 342.6 Cray CS6400, 4 cpu 8 188.9 184.3 188.4 188.8 Cray CS6400, 1 cpu 8 51.1 49.9 50.0 50.2 CM-5E 32 PEs 8 11759.9 11759.1 12383.9 12378.3 CM-5E 32 PEs 4 8573.8 7055.2 8043.8 8036.5 CM-200 64k 10 (est) 8 42011.2 42010.4 47451.2 45225.6 CM-200 8k 10 MHz 8 5251.4 5251.3 5931.4 5653.2 CM-2 64k 8 MHz (est) 8 29465.9 30592.8 33520.0 32000.0 CM-2 64k 7 MHz (est) 8 25796.0 25796.0 29259.2 27958.4 CM-2 8k 8 MHz 8 3683.2 3824.1 4190.0 4000.0 CM-2 8k 7 MHz 8 3224.5 3224.5 3657.4 3494.8 NEC SX3/44 1 cpu 8 16941.0 15640.7 22436.5 21972.2 NEC SX3/12 1 cpu 8 8302.2 7670.9 9712.8 9712.9 NEC SX3/12 1 cpu 4 4151.5 3835.5 4993.8 4993.9 ETA-10E 8 2974.7 3010.5 4357.5 4214.8 IBM 3090J/VF 1 cpu 8 251.9 250.1 288.5 264.6 IBM 3090J/VF 1 cpu 4 202.5 166.5 173.0 173.9 IBM 3090J scalar 8 117.5 95.5 132.5 104.7 IBM 3090J scalar 4 92.8 56.1 72.6 68.5 Convex C3420 2 cpu 8 289.1 273.1 302.7 300.3 Convex C3420 2 cpu 4 242.7 238.8 263.7 260.0 Convex C3410 1 cpu 8 177.6 172.5 179.0 178.4 Convex C3410 1 cpu 4 155.5 150.9 157.1 154.4 Convex C3220 2 cpu 8 334.3 333.0 337.3 336.5 Convex C3220 2 cpu 4 178.7 177.5 211.0 208.8 Convex C3210 1 cpu 8 180.1 179.4 181.4 180.8 Convex C3210 1 cpu 4 93.4 92.2 108.9 109.0 FPS MCP728 7 cpu 8 1040.5 1036.8 890.4 887.6 FPS MCP728 7 cpu 4 1046.2 478.0 393.0 394.8 FPS MCP104 1 cpu 8 154.1 154.0 131.2 130.9 FPS MCP104 1 cpu 4 154.6 103.7 115.9 93.1 FPS 511 SPARC vector 8 250.5 170.8 190.5 190.2 FPS 511 SPARC vector 4 124.6 85.3 95.0 95.0 FPS 511 SPARC scalar 8 47.9 45.0 36.5 34.2 FPS 511 SPARC scalar 4 37.2 31.0 27.5 26.6 FPS 511 EA vector 8 246.2 168.4 187.0 187.0 FPS 511 EA vector 4 121.2 83.3 93.0 93.0 FPS 510 EA scalar 8 26.1 21.6 23.7 21.4 FPS 510 EA scalar 4 14.1 11.6 13.1 11.8 FPS 511 (1 vector) 8 288.0 180.0 216.0 196.4 Meiko CS-2 1 cpu 8 621.3 621.3 610.3 610.3 MIPS RC6280 8 53.5 58.2 56.5 56.5 Convex C1XP 8 71.3 71.0 71.3 71.3 Convex C1XP 4 62.9 62.5 64.4 64.5 Stardent Vistra 800b 8 147.7 147.7 115.2 115.2 Stardent ST2000 8 163.8 163.8 179.8 179.8 Stardent P3 3 cpu 8 228.6 246.2 246.2 200.0 Stardent P3 8 83.1 78.0 78.0 70.1 IBM RS6000-990 8 663.4 533.4 714.5 713.8 IBM RS6000-990/128 8 333.3 333.3 363.6 363.6 IBM RS6000-590 8 600.0 533.3 654.5 654.5 IBM RS6000-580 8 275.9 275.9 250.0 240.0 IBM RS6000-580 4 173.9 142.9 187.5 166.7 IBM RS6000-560 8 228.6 228.6 240.0 240.0 IBM RS6000-550 8 240.0 240.0 196.4 216.0 IBM RS6000-950 8 193.9 177.8 195.9 192.0 IBM RS6000-950 4 123.1 74.4 102.1 100.0 IBM RS6000-540 8 145.5 129.7 144.0 144.0 IBM RS6000-530 8 114.3 106.7 109.1 114.3 IBM RS6000-530 4 72.7 44.4 61.5 60.0 IBM RS6000-355 8 133.3 133.3 114.3 120.0 IBM RS6000-355 4 100.0 84.2 85.7 85.7 IBM RS6000-320H 8 96.0 96.0 90.0 90.0 IBM RS6000-320H 4 60.0 40.0 54.0 54.0 IBM RS6000-320 8 61.5 61.5 60.0 60.0 IBM RS6000-320 4 44.4 33.3 40.0 40.0 IBM RS/6000-250 8 71.1 67.4 73.8 73.8 IBM RS/6000-250 4 58.2 51.2 53.3 53.3 HP 9000/755 8 68.6 68.6 72.0 80.0 HP 9000/730 8 53.3 48.0 55.4 55.4 HP 9000/730 4 40.0 36.9 45.0 40.0 HP 9000/720 8 48.5 50.0 50.0 53.3 HP 9000/720 4 44.4 47.1 48.0 52.1 Alliant FX/2800 14 cpu 8 309.0 305.9 297.8 292.0 Alliant FX/80 8 cpu 8 72.8 71.6 76.3 76.5 Alliant VFX80 3 cpu 8 41.7 41.6 43.1 43.0 Alliant VFX80 3 cpu 4 23.8 23.8 25.2 25.3 Apollo DN10010 8 41.0 48.0 54.0 54.0 DEC 7000/610 4 cpu 8 261.9 257.4 261.1 265.9 DEC 7000/610 2 cpu 8 163.9 166.0 161.8 157.1 DEC 7000/610 1 cpu 8 91.6 91.6 88.8 86.9 DEC 2100 A500 2 cpu 8 133.3 133.6 129.6 127.7 DEC 2100 A500 1 cpu 8 79.2 80.8 79.1 79.3 VAX 6000-410 vector 8 58.5 58.5 61.4 51.2 VAX 9000/420 1 cpu 8 144.0 164.6 172.8 157.1 VAX 9000/420 1 cpu 4 82.3 82.3 78.5 78.5 DEC 4000/710 8 84.6 82.3 85.1 83.8 DEC 4000/710 4 80.4 80.4 79.2 78.8 DEC 3000/500 8 100.4 96.4 98.4 99.7 DEC 3000/300 8 33.4 33.5 39.6 38.9 DEC 5000/200 8 28.6 24.6 24.3 23.0 DEC 5000/200 4 24.4 20.3 21.6 18.3 Omron Luna88k 8 14.4 14.4 16.0 13.1 Omron Luna88k 4 14.1 13.3 15.6 13.3 SGI Power Challenge, 8 cpu 8 661.0 578.9 660.8 749.3 SGI Power Challenge, 6 cpu 8 571.4 538.7 602.0 647.1 SGI Power Challenge, 4 cpu 8 439.8 407.8 463.1 478.8 SGI Power Challenge, 2 cpu 8 234.8 222.8 252.3 254.8 SGI Pow Chal 8 banks 8 134.9 134.8 134.8 134.8 SGI Pow Chal 2 banks 8 131.0 131.9 129.3 129.1 SGI Challenge 150 MHz 8 58.2 57.1 57.1 62.3 SGI Challenge 150 MHz 4 66.7 66.7 66.7 63.2 SGI Challenge 1 cpu 8 57.1 55.2 53.3 54.5 SGI Challenge 1 cpu 4 53.3 47.1 50.0 48.0 SGI Crimson 8 61.5 59.3 58.5 60.0 SGI Crimson 4 57.1 53.5 54.5 52.2 SGI 4D/240 8 18.4 16.5 19.8 19.0 SGI 4D/240 4 18.8 14.4 18.3 15.8 SGI 4D/35 8 53.5 36.9 48.0 42.4 SGI 4D/35 4 50.0 28.6 37.5 25.0 SGI Indigo 8 36.9 34.3 36.0 36.0 SGI Indigo 4 28.6 21.1 26.1 26.7 SGI 4D/25 8 12.7 10.1 10.4 10.0 SGI 4D/25 4 12.5 9.4 10.1 9.6 Sun SparcClassic 8 57.6 48.0 48.0 43.2 Sun SparcCenter 2000 8 34.4 36.7 35.2 35.0 Sun SS10/41 1 cpu 8 48.0 48.0 54.0 54.0 Sun SS10/41 2 cpu 8 48.0 48.0 48.0 48.0 Sun SS10/41 4 34.3 38.4 36.9 37.9 Sun SS10/30 8 42.1 46.2 46.2 46.2 Sun SS10/30 4 33.9 41.7 39.0 34.1 Sun 670 8 28.2 32.0 30.0 30.0 Sun 670 4 21.1 21.1 22.2 24.0 Sun 4/490 8 25.0 25.8 25.5 24.5 Sun 4/490 4 25.8 14.0 23.1 16.2 Sun SS2 (4/75) 8 30.0 30.0 28.8 26.7 Sun SS2 (4/75) 4 19.0 17.4 19.4 20.7 Sun SS1 8 14.1 12.3 13.6 13.1 Sun SS1 4 12.9 8.7 12.2 8.3 NeXTStation 68040 8 14.2 11.5 16.4 16.1 NeXTStation 68040 4 13.4 7.7 10.5 8.4 Dell 486 DX/2-66 8 33.3 16.5 22.0 18.8 Intel Pentium/60 8 37.2 62.1 61.3 58.5