------------------------------------------------------------- STREAM Memory Benchmark ------------------------------------------------------------- The Test will run some minutes please be patient. Total memory required = 160.0 MB. Each test is run 3 times, but only the *best* time for each is used. ------------------------------------------------------------- Memory throughput Working on Arrays of 80 MB. ------------------------------------------------------------- Read test (summing up the array). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time read 8 537.4143 0.1491 0.1489 0.1494 read 32 2327.0661 0.0344 0.0344 0.0344 read 64 2233.6415 0.0358 0.0358 0.0358 read 32x2 2499.8459 0.0322 0.0320 0.0323 read 32 CP3 2226.0545 0.0362 0.0359 0.0364 read 32 CP4 2228.3607 0.0360 0.0359 0.0360 read 32 CP5 2233.1955 0.0359 0.0358 0.0359 read 32 CP6 2227.1035 0.0360 0.0359 0.0360 read 32x4 CP3 2636.6212 0.0304 0.0303 0.0304 read 32x4 CP4 2614.8824 0.0306 0.0306 0.0307 read 32x4 CP5 2604.6724 0.0307 0.0307 0.0307 read 32x4 CP6 2606.5339 0.0307 0.0307 0.0308 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 1493.0134 0.0536 0.0536 0.0536 write 32 1441.1063 0.0556 0.0555 0.0556 write 64 1572.9771 0.0516 0.0509 0.0523 write 32x2 1436.6822 0.0560 0.0557 0.0562 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 1011.7179 0.1582 0.1581 0.1582 cmp 32 2858.6158 0.0562 0.0560 0.0564 cmp 64 2963.1254 0.0540 0.0540 0.0540 cmp 32x2 2885.7449 0.0555 0.0554 0.0555 cmp 32 CP2 2656.1778 0.0606 0.0602 0.0609 cmp 32 CP3 2662.2789 0.0602 0.0601 0.0602 cmp 32 CP4 2661.6560 0.0602 0.0601 0.0602 cmp 32 CP5 2656.2094 0.0603 0.0602 0.0603 cmp 32 CP6 2640.4388 0.0608 0.0606 0.0611 cmp 32x4 CP2 2699.4716 0.0594 0.0593 0.0595 cmp 32x4 CP3 2709.7066 0.0591 0.0590 0.0591 cmp 32x4 CP4 2711.7213 0.0593 0.0590 0.0595 cmp 32x4 CP5 2713.3330 0.0590 0.0590 0.0591 cmp 32x4 CP6 2712.2363 0.0593 0.0590 0.0597 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 1895.8699 0.0847 0.0844 0.0849 copy 32 2164.7043 0.0742 0.0739 0.0746 copy 64 2223.9520 0.0720 0.0719 0.0720 copy 32x2 2166.1716 0.0741 0.0739 0.0743 copy 32 CP2 2166.8781 0.0739 0.0738 0.0740 copy 32 CP3 2168.1942 0.0744 0.0738 0.0749 copy 32 CP4 2138.9552 0.0748 0.0748 0.0748 copy 32 CP5 2132.2000 0.0754 0.0750 0.0757 copy 32x4 CP2 2216.9578 0.0726 0.0722 0.0731 copy 32x4 CP3 2200.9762 0.0729 0.0727 0.0731 copy 32x4 CP4 2173.9748 0.0740 0.0736 0.0745 copy 32x4 CP5 2188.6656 0.0732 0.0731 0.0732 copy 64x4 CP4 2225.6853 0.0720 0.0719 0.0721 glibcb memcpy 2155.5194 0.0743 0.0742 0.0744 bmove512 2165.1164 0.0742 0.0739 0.0744 FC64 2257.4371 0.0710 0.0709 0.0711 ------------------------------------------------------------- 2nd level cache throughput Working on Arrays of 80 KB. ------------------------------------------------------------- Read test (summing up the array). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time read 8 574.3370 0.1395 0.1393 0.1397 read 32 4549.0750 0.0176 0.0176 0.0176 read 64 3065.7316 0.0261 0.0261 0.0262 read 32x2 4589.2679 0.0174 0.0174 0.0174 read 32 CP3 3061.0330 0.0261 0.0261 0.0262 read 32 CP4 3060.4467 0.0262 0.0261 0.0263 read 32 CP5 3063.4644 0.0263 0.0261 0.0264 read 32 CP6 3061.2844 0.0262 0.0261 0.0262 read 32x4 CP3 10317.1393 0.0078 0.0078 0.0078 read 32x4 CP4 10354.7082 0.0078 0.0077 0.0078 read 32x4 CP5 10353.4302 0.0078 0.0077 0.0078 read 32x4 CP6 10388.3690 0.0078 0.0077 0.0079 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 2296.6128 0.0348 0.0348 0.0349 write 32 9182.6803 0.0087 0.0087 0.0087 write 64 18339.7639 0.0044 0.0044 0.0044 write 32x2 9187.9606 0.0087 0.0087 0.0087 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 1147.6525 0.1396 0.1394 0.1397 cmp 32 7439.7597 0.0218 0.0215 0.0220 cmp 64 6120.6702 0.0262 0.0261 0.0262 cmp 32x2 7950.7220 0.0202 0.0201 0.0203 cmp 32 CP2 4487.9566 0.0358 0.0357 0.0359 cmp 32 CP3 4473.5067 0.0358 0.0358 0.0358 cmp 32 CP4 4468.0265 0.0359 0.0358 0.0360 cmp 32 CP5 4450.4850 0.0362 0.0360 0.0364 cmp 32 CP6 4472.5229 0.0359 0.0358 0.0360 cmp 32x4 CP2 4596.7494 0.0348 0.0348 0.0348 cmp 32x4 CP3 4597.1588 0.0348 0.0348 0.0349 cmp 32x4 CP4 4584.1267 0.0351 0.0349 0.0352 cmp 32x4 CP5 4593.0685 0.0349 0.0348 0.0349 cmp 32x4 CP6 4591.3716 0.0353 0.0348 0.0357 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 2607.9419 0.0614 0.0614 0.0615 copy 32 10511.8754 0.0153 0.0152 0.0153 copy 64 21724.4065 0.0074 0.0074 0.0074 copy 32x2 15150.0957 0.0106 0.0106 0.0106 copy 32 CP2 6107.3018 0.0262 0.0262 0.0262 copy 32 CP3 6097.8123 0.0263 0.0262 0.0263 copy 32 CP4 6046.6607 0.0265 0.0265 0.0266 copy 32 CP5 6111.3062 0.0263 0.0262 0.0264 copy 32x4 CP2 13914.0520 0.0115 0.0115 0.0115 copy 32x4 CP3 13887.5616 0.0115 0.0115 0.0115 copy 32x4 CP4 13912.0328 0.0115 0.0115 0.0115 copy 32x4 CP5 13896.1887 0.0115 0.0115 0.0115 copy 64x4 CP4 25268.7943 0.0063 0.0063 0.0063 glibcb memcpy 15637.2598 0.0102 0.0102 0.0103 bmove512 13177.4626 0.0122 0.0121 0.0122 FC64 27285.5719 0.0059 0.0059 0.0059 ------------------------------------------------------------- 1st level cache throughput Working on Arrays of 800 B. ------------------------------------------------------------- Read test (summing up the array). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time read 8 574.2298 0.1394 0.1393 0.1394 read 32 4414.7664 0.0181 0.0181 0.0181 read 64 3058.1038 0.0262 0.0262 0.0263 read 32x2 4546.4862 0.0176 0.0176 0.0177 read 32 CP3 2900.3243 0.0277 0.0276 0.0277 read 32 CP4 2882.0642 0.0278 0.0278 0.0279 read 32 CP5 2893.5214 0.0278 0.0276 0.0279 read 32 CP6 2907.3860 0.0275 0.0275 0.0276 read 32x4 CP3 10655.2450 0.0075 0.0075 0.0075 read 32x4 CP4 10662.3553 0.0075 0.0075 0.0075 read 32x4 CP5 10662.3553 0.0075 0.0075 0.0075 read 32x4 CP6 10662.3553 0.0076 0.0075 0.0077 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 2229.2193 0.0360 0.0359 0.0361 write 32 8032.9492 0.0100 0.0100 0.0100 write 64 15519.3710 0.0052 0.0052 0.0052 write 32x2 8545.1988 0.0094 0.0094 0.0094 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 1142.6208 0.1404 0.1400 0.1407 cmp 32 7349.5635 0.0220 0.0218 0.0223 cmp 64 5882.9753 0.0272 0.0272 0.0272 cmp 32x2 7945.9209 0.0202 0.0201 0.0203 cmp 32 CP2 4487.8366 0.0360 0.0357 0.0363 cmp 32 CP3 4464.2814 0.0361 0.0358 0.0363 cmp 32 CP4 4448.7739 0.0361 0.0360 0.0363 cmp 32 CP5 4490.9900 0.0359 0.0356 0.0362 cmp 32 CP6 4452.2271 0.0363 0.0359 0.0366 cmp 32x4 CP2 4563.4904 0.0351 0.0351 0.0351 cmp 32x4 CP3 4566.3471 0.0350 0.0350 0.0351 cmp 32x4 CP4 4566.9997 0.0351 0.0350 0.0351 cmp 32x4 CP5 4569.2075 0.0351 0.0350 0.0352 cmp 32x4 CP6 4561.5362 0.0352 0.0351 0.0354 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 2565.1274 0.0626 0.0624 0.0629 copy 32 9905.2212 0.0162 0.0162 0.0163 copy 64 21638.9462 0.0074 0.0074 0.0074 copy 32x2 14598.7217 0.0110 0.0110 0.0110 copy 32 CP2 5879.3675 0.0272 0.0272 0.0272 copy 32 CP3 5909.2911 0.0271 0.0271 0.0271 copy 32 CP4 5900.4066 0.0273 0.0271 0.0275 copy 32 CP5 5907.1064 0.0271 0.0271 0.0271 copy 32x4 CP2 13546.6731 0.0118 0.0118 0.0118 copy 32x4 CP3 13437.3601 0.0119 0.0119 0.0119 copy 32x4 CP4 13559.2637 0.0118 0.0118 0.0118 copy 32x4 CP5 13520.1999 0.0118 0.0118 0.0119 copy 64x4 CP4 24968.8819 0.0064 0.0064 0.0064 glibcb memcpy 13518.0211 0.0120 0.0118 0.0122 bmove512 9424.4757 0.0171 0.0170 0.0172 FC64 29080.4108 0.0055 0.0055 0.0055 -------------------------------------------------------------