------------------------------------------------------------- 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 8 195.0217 0.4110 0.4102 0.4118 read 32 242.0363 0.3306 0.3305 0.3308 read 64 261.4404 0.3061 0.3060 0.3062 read 32x2 261.6979 0.3059 0.3057 0.3061 read 32 CP3 485.6669 0.1649 0.1647 0.1651 read 32 CP4 587.7678 0.1361 0.1361 0.1362 read 32 CP5 657.5806 0.1217 0.1217 0.1217 read 32 CP6 649.1449 0.1233 0.1232 0.1234 read 32x4 CP3 544.1212 0.1471 0.1470 0.1472 read 32x4 CP4 642.5553 0.1246 0.1245 0.1248 read 32x4 CP5 678.7421 0.1180 0.1179 0.1182 read 32x4 CP6 652.1451 0.1228 0.1227 0.1229 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 251.7046 0.3178 0.3178 0.3179 write 32 493.0661 0.1623 0.1623 0.1623 write 64 492.2531 0.1626 0.1625 0.1626 write 32x2 492.3289 0.1629 0.1625 0.1633 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 236.7929 0.6760 0.6757 0.6764 cmp 32 275.3721 0.5813 0.5810 0.5816 cmp 64 276.3179 0.5792 0.5790 0.5793 cmp 32x2 275.3796 0.5814 0.5810 0.5817 cmp 32 CP2 436.2775 0.3669 0.3667 0.3671 cmp 32 CP3 646.1959 0.2479 0.2476 0.2483 cmp 32 CP4 634.8754 0.2522 0.2520 0.2523 cmp 32 CP5 642.7203 0.2491 0.2489 0.2492 cmp 32 CP6 631.1388 0.2537 0.2535 0.2539 cmp 32x4 CP2 622.8467 0.2570 0.2569 0.2572 cmp 32x4 CP3 669.2489 0.2396 0.2391 0.2401 cmp 32x4 CP4 632.0959 0.2533 0.2531 0.2535 cmp 32x4 CP5 654.5897 0.2445 0.2444 0.2446 cmp 32x4 CP6 618.3171 0.2590 0.2588 0.2593 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 336.5999 0.4756 0.4753 0.4758 copy 32 379.7769 0.4215 0.4213 0.4216 copy 64 391.2113 0.4092 0.4090 0.4093 copy 32x2 397.8842 0.4024 0.4021 0.4027 copy 32 CP2 451.6634 0.3543 0.3542 0.3543 copy 32 CP3 515.4805 0.3104 0.3104 0.3105 copy 32 CP4 573.3659 0.2791 0.2791 0.2791 copy 32 CP5 629.4599 0.2542 0.2542 0.2542 copy 32x4 CP2 503.6690 0.3179 0.3177 0.3182 copy 32x4 CP3 530.8087 0.3040 0.3014 0.3066 copy 32x4 CP4 597.5814 0.2678 0.2677 0.2679 copy 32x4 CP5 612.9559 0.2611 0.2610 0.2612 copy 64x4 CP4 529.8222 0.3021 0.3020 0.3023 copy 64x4 CP4C 554.8654 0.2885 0.2884 0.2886 glibcb memcpy 375.4653 0.4263 0.4261 0.4265 bmove512 380.4224 0.4208 0.4206 0.4210 FC64 514.4314 0.3111 0.3110 0.3111 ------------------------------------------------------------- 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 8 436.1480 0.1834 0.1834 0.1834 read 32 1769.0487 0.0453 0.0452 0.0453 read 64 1594.4514 0.0502 0.0502 0.0502 read 32x2 1873.0529 0.0427 0.0427 0.0427 read 32 CP3 1328.8410 0.0602 0.0602 0.0602 read 32 CP4 1327.9575 0.0602 0.0602 0.0602 read 32 CP5 1327.6475 0.0603 0.0603 0.0603 read 32 CP6 1328.8410 0.0602 0.0602 0.0602 read 32x4 CP3 2274.7842 0.0352 0.0352 0.0352 read 32x4 CP4 2277.8418 0.0351 0.0351 0.0351 read 32x4 CP5 2277.3316 0.0351 0.0351 0.0352 read 32x4 CP6 2273.5666 0.0352 0.0352 0.0352 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 678.3634 0.1179 0.1179 0.1179 write 32 1673.2957 0.0478 0.0478 0.0478 write 64 2124.3839 0.0377 0.0377 0.0377 write 32x2 1677.8474 0.0477 0.0477 0.0477 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 612.8389 0.2613 0.2611 0.2614 cmp 32 2276.1267 0.0703 0.0703 0.0704 cmp 64 2649.6651 0.0604 0.0604 0.0604 cmp 32x2 2273.8593 0.0704 0.0704 0.0704 cmp 32 CP2 1677.5371 0.0954 0.0954 0.0955 cmp 32 CP3 1676.5857 0.0955 0.0954 0.0956 cmp 32 CP4 1676.9921 0.0954 0.0954 0.0954 cmp 32 CP5 1676.6569 0.0954 0.0954 0.0955 cmp 32 CP6 1676.6402 0.0957 0.0954 0.0961 cmp 32x4 CP2 2655.8309 0.0603 0.0602 0.0603 cmp 32x4 CP3 2655.2950 0.0603 0.0603 0.0604 cmp 32x4 CP4 2649.0899 0.0604 0.0604 0.0604 cmp 32x4 CP5 2652.2622 0.0604 0.0603 0.0605 cmp 32x4 CP6 2650.3244 0.0604 0.0604 0.0605 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 637.3233 0.2511 0.2510 0.2511 copy 32 2124.6126 0.0753 0.0753 0.0754 copy 64 3546.3241 0.0451 0.0451 0.0451 copy 32x2 2009.4700 0.0796 0.0796 0.0797 copy 32 CP2 1692.4032 0.0947 0.0945 0.0949 copy 32 CP3 1687.3352 0.0949 0.0948 0.0949 copy 32 CP4 1686.6779 0.0949 0.0949 0.0949 copy 32 CP5 1692.0661 0.0947 0.0946 0.0948 copy 32x4 CP2 2148.6594 0.0746 0.0745 0.0748 copy 32x4 CP3 2147.2775 0.0746 0.0745 0.0748 copy 32x4 CP4 2147.9992 0.0746 0.0745 0.0747 copy 32x4 CP5 2149.2374 0.0744 0.0744 0.0745 copy 64x4 CP4 3414.3406 0.0469 0.0469 0.0469 copy 64x4 CP4C 3392.6264 0.0472 0.0472 0.0472 glibcb memcpy 2118.7032 0.0756 0.0755 0.0757 bmove512 2121.5096 0.0755 0.0754 0.0756 FC64 3429.9415 0.0467 0.0466 0.0467 ------------------------------------------------------------- 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 8 477.5867 0.1676 0.1675 0.1678 read 32 2931.4910 0.0273 0.0273 0.0273 read 64 1551.6213 0.0516 0.0516 0.0516 read 32x2 3142.0655 0.0255 0.0255 0.0255 read 32 CP3 1310.7661 0.0610 0.0610 0.0610 read 32 CP4 1310.6176 0.0611 0.0610 0.0611 read 32 CP5 1310.4692 0.0611 0.0610 0.0612 read 32 CP6 1310.5511 0.0610 0.0610 0.0611 read 32x4 CP3 3095.1418 0.0259 0.0258 0.0259 read 32x4 CP4 3094.5423 0.0301 0.0259 0.0344 read 32x4 CP5 3096.9415 0.0261 0.0258 0.0264 read 32x4 CP6 3099.2012 0.0259 0.0258 0.0261 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 842.3798 0.0974 0.0950 0.0998 write 32 3242.0368 0.0272 0.0247 0.0297 write 64 2629.9462 0.0305 0.0304 0.0306 write 32x2 3284.8839 0.0244 0.0244 0.0244 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 657.2084 0.2435 0.2435 0.2435 cmp 32 3042.5752 0.0526 0.0526 0.0526 cmp 64 2553.4251 0.0629 0.0627 0.0632 cmp 32x2 3096.4556 0.0517 0.0517 0.0517 cmp 32 CP2 1966.9809 0.0814 0.0813 0.0814 cmp 32 CP3 1967.8749 0.0813 0.0813 0.0814 cmp 32 CP4 1967.8461 0.0813 0.0813 0.0813 cmp 32 CP5 1968.1404 0.0813 0.0813 0.0813 cmp 32 CP6 1967.9442 0.0813 0.0813 0.0813 cmp 32x4 CP2 2611.1789 0.0613 0.0613 0.0613 cmp 32x4 CP3 2613.0191 0.0613 0.0612 0.0614 cmp 32x4 CP4 2612.6733 0.0613 0.0612 0.0613 cmp 32x4 CP5 2610.7522 0.0613 0.0613 0.0613 cmp 32x4 CP6 2611.9919 0.0614 0.0613 0.0615 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 662.6439 0.2415 0.2415 0.2416 copy 32 2640.0025 0.0606 0.0606 0.0606 copy 64 5128.7258 0.0312 0.0312 0.0312 copy 32x2 2661.4765 0.0601 0.0601 0.0601 copy 32 CP2 1601.5059 0.0999 0.0999 0.1000 copy 32 CP3 1601.0245 0.0999 0.0999 0.0999 copy 32 CP4 1601.2499 0.1000 0.0999 0.1000 copy 32 CP5 1601.1544 0.0999 0.0999 0.0999 copy 32x4 CP2 3157.9453 0.0507 0.0507 0.0508 copy 32x4 CP3 3156.4303 0.0508 0.0507 0.0509 copy 32x4 CP4 3155.0205 0.0508 0.0507 0.0508 copy 32x4 CP5 3157.8710 0.0507 0.0507 0.0508 copy 64x4 CP4 5126.0619 0.0312 0.0312 0.0312 copy 64x4 CP4C 4172.7362 0.0384 0.0383 0.0384 glibcb memcpy 2491.8631 0.0642 0.0642 0.0642 bmove512 2060.0201 0.0777 0.0777 0.0778 FC64 4635.5184 0.0346 0.0345 0.0347 -------------------------------------------------------------