------------------------------------------------------------- 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 8 152.8292 0.5271 0.5235 0.5307 read 32 202.2843 0.3971 0.3955 0.3987 read 64 207.5254 0.3931 0.3855 0.4007 read 32x2 203.0829 0.4024 0.3939 0.4109 read 32 CP3 259.5547 0.3092 0.3082 0.3102 read 32 CP4 259.8408 0.3181 0.3079 0.3283 read 32 CP5 260.0881 0.3082 0.3076 0.3087 read 32 CP6 257.9432 0.3103 0.3101 0.3104 read 32x4 CP3 260.0044 0.3126 0.3077 0.3174 read 32x4 CP4 258.1078 0.3126 0.3099 0.3152 read 32x4 CP5 259.0212 0.3166 0.3089 0.3243 read 32x4 CP6 257.6083 0.3109 0.3105 0.3112 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 247.9266 0.3288 0.3227 0.3349 write 32 314.7598 0.2542 0.2542 0.2542 write 64 601.6811 0.1331 0.1330 0.1332 write 32x2 318.2889 0.2544 0.2513 0.2575 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 171.8412 0.9334 0.9311 0.9357 cmp 32 209.0586 0.7665 0.7653 0.7676 cmp 64 206.6345 0.7755 0.7743 0.7767 cmp 32x2 207.9794 0.7745 0.7693 0.7797 cmp 32 CP2 249.7145 0.6469 0.6407 0.6530 cmp 32 CP3 254.8420 0.6282 0.6278 0.6286 cmp 32 CP4 254.3433 0.6332 0.6291 0.6373 cmp 32 CP5 255.5636 0.6359 0.6261 0.6458 cmp 32 CP6 260.4498 0.6217 0.6143 0.6291 cmp 32x4 CP2 261.0656 0.6262 0.6129 0.6395 cmp 32x4 CP3 261.1196 0.6198 0.6127 0.6269 cmp 32x4 CP4 256.4378 0.6373 0.6239 0.6507 cmp 32x4 CP5 262.8501 0.6198 0.6087 0.6308 cmp 32x4 CP6 260.8243 0.6230 0.6134 0.6325 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 171.0762 0.9426 0.9353 0.9499 copy 32 150.6984 1.0649 1.0617 1.0680 copy 64 148.9062 1.0779 1.0745 1.0814 copy 32x2 179.3266 0.8952 0.8922 0.8981 copy 32 CP2 210.6469 0.7626 0.7596 0.7656 copy 32 CP3 232.5899 0.6882 0.6879 0.6886 copy 32 CP4 239.6781 0.6683 0.6676 0.6690 copy 32 CP5 245.9991 0.6540 0.6504 0.6576 copy 32x4 CP2 227.1685 0.7068 0.7043 0.7092 copy 32x4 CP3 242.7936 0.6608 0.6590 0.6625 copy 32x4 CP4 258.3599 0.6241 0.6193 0.6289 copy 32x4 CP5 262.3518 0.6111 0.6099 0.6124 copy 64x4 CP4 276.4841 0.5822 0.5787 0.5856 glibcb memcpy 186.8131 0.8634 0.8565 0.8704 bmove512 150.6112 1.0634 1.0623 1.0645 FC64 252.6979 0.6434 0.6332 0.6536 memcpy 750 327.1581 0.4930 0.4891 0.4970 ------------------------------------------------------------- 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 8 260.6356 0.3142 0.3069 0.3214 read 32 1259.3286 0.0637 0.0635 0.0639 read 64 1262.8026 0.0635 0.0634 0.0636 read 32x2 1332.7785 0.0601 0.0600 0.0602 read 32 CP3 1206.5469 0.0664 0.0663 0.0665 read 32 CP4 1204.3513 0.0665 0.0664 0.0665 read 32 CP5 1204.7275 0.0672 0.0664 0.0679 read 32 CP6 1202.4265 0.0672 0.0665 0.0678 read 32x4 CP3 1797.5460 0.0447 0.0445 0.0448 read 32x4 CP4 1812.7830 0.0442 0.0441 0.0442 read 32x4 CP5 1796.0162 0.0447 0.0445 0.0448 read 32x4 CP6 1810.4844 0.0442 0.0442 0.0442 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 393.2867 0.2057 0.2034 0.2080 write 32 1325.8219 0.0603 0.0603 0.0603 write 64 1684.9585 0.0475 0.0475 0.0476 write 32x2 1331.0286 0.0608 0.0601 0.0615 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 393.8364 0.4065 0.4063 0.4067 cmp 32 1563.0777 0.1025 0.1024 0.1026 cmp 64 1612.6589 0.1024 0.0992 0.1055 cmp 32x2 1622.1429 0.0991 0.0986 0.0996 cmp 32 CP2 1323.9988 0.1213 0.1208 0.1217 cmp 32 CP3 1322.6210 0.1219 0.1210 0.1229 cmp 32 CP4 1326.4011 0.1207 0.1206 0.1209 cmp 32 CP5 1324.6025 0.1216 0.1208 0.1224 cmp 32 CP6 1331.3692 0.1206 0.1202 0.1210 cmp 32x4 CP2 2094.5731 0.0764 0.0764 0.0764 cmp 32x4 CP3 2096.8506 0.0765 0.0763 0.0768 cmp 32x4 CP4 2099.3235 0.0768 0.0762 0.0775 cmp 32x4 CP5 2101.4205 0.0793 0.0761 0.0824 cmp 32x4 CP6 2087.7376 0.0770 0.0766 0.0773 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 503.9974 0.3181 0.3175 0.3188 copy 32 1685.5933 0.0954 0.0949 0.0959 copy 64 2841.7656 0.0563 0.0563 0.0564 copy 32x2 1626.4951 0.0990 0.0984 0.0997 copy 32 CP2 1316.2859 0.1227 0.1216 0.1238 copy 32 CP3 1326.1940 0.1278 0.1206 0.1350 copy 32 CP4 1302.2629 0.1231 0.1229 0.1233 copy 32 CP5 1311.1963 0.1262 0.1220 0.1303 copy 32x4 CP2 1757.0249 0.0920 0.0911 0.0929 copy 32x4 CP3 1752.4459 0.0917 0.0913 0.0921 copy 32x4 CP4 1734.1195 0.0924 0.0923 0.0925 copy 32x4 CP5 1768.5033 0.0955 0.0905 0.1005 copy 64x4 CP4 3077.5836 0.0535 0.0520 0.0551 glibcb memcpy 1662.0608 0.0972 0.0963 0.0982 bmove512 1675.0206 0.0993 0.0955 0.1030 FC64 2713.4317 0.0606 0.0590 0.0623 memcpy 750 1734.4198 0.0925 0.0922 0.0928 ------------------------------------------------------------- 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 8 262.7421 0.3047 0.3045 0.3049 read 32 1568.9392 0.0512 0.0510 0.0513 read 64 1234.8946 0.0648 0.0648 0.0648 read 32x2 2097.7551 0.0381 0.0381 0.0382 read 32 CP3 1542.4063 0.0519 0.0519 0.0520 read 32 CP4 1542.4985 0.0520 0.0519 0.0521 read 32 CP5 1544.8207 0.0519 0.0518 0.0520 read 32 CP6 1543.6267 0.0520 0.0518 0.0522 read 32x4 CP3 2453.9933 0.0327 0.0326 0.0327 read 32x4 CP4 2457.0841 0.0341 0.0326 0.0357 read 32x4 CP5 2454.2805 0.0351 0.0326 0.0376 read 32x4 CP6 2460.5616 0.0327 0.0325 0.0328 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 395.2551 0.2033 0.2024 0.2043 write 32 1568.8438 0.0511 0.0510 0.0512 write 64 2088.8228 0.0383 0.0383 0.0383 write 32x2 2664.6151 0.0302 0.0300 0.0304 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 430.7165 0.3717 0.3715 0.3718 cmp 32 2088.6083 0.0767 0.0766 0.0768 cmp 64 1929.0365 0.0868 0.0829 0.0906 cmp 32x2 2288.5225 0.0700 0.0699 0.0701 cmp 32 CP2 1562.7610 0.1034 0.1024 0.1044 cmp 32 CP3 1564.1670 0.1037 0.1023 0.1050 cmp 32 CP4 1557.1622 0.1033 0.1028 0.1039 cmp 32 CP5 1558.9854 0.1028 0.1026 0.1030 cmp 32 CP6 1563.2342 0.1024 0.1024 0.1025 cmp 32x4 CP2 2252.9506 0.0710 0.0710 0.0711 cmp 32x4 CP3 2250.1329 0.0712 0.0711 0.0712 cmp 32x4 CP4 2252.3154 0.0711 0.0710 0.0712 cmp 32x4 CP5 2242.4037 0.0719 0.0714 0.0725 cmp 32x4 CP6 2249.4315 0.0713 0.0711 0.0715 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 526.7956 0.3069 0.3037 0.3100 copy 32 2094.1025 0.0815 0.0764 0.0867 copy 64 4155.1932 0.0386 0.0385 0.0388 copy 32x2 1913.0289 0.0837 0.0836 0.0837 copy 32 CP2 1277.2739 0.1284 0.1253 0.1315 copy 32 CP3 1271.0012 0.1259 0.1259 0.1260 copy 32 CP4 1275.8970 0.1257 0.1254 0.1260 copy 32 CP5 1276.3241 0.1254 0.1254 0.1254 copy 32x4 CP2 2468.4810 0.0649 0.0648 0.0649 copy 32x4 CP3 2473.0293 0.0648 0.0647 0.0649 copy 32x4 CP4 2455.6456 0.0661 0.0652 0.0669 copy 32x4 CP5 2469.5529 0.0653 0.0648 0.0658 copy 64x4 CP4 4067.7462 0.0395 0.0393 0.0398 glibcb memcpy 1985.5810 0.0806 0.0806 0.0806 bmove512 1632.7515 0.0980 0.0980 0.0981 FC64 3557.5286 0.0455 0.0450 0.0461 memcpy 750 1763.6650 0.0930 0.0907 0.0953 -------------------------------------------------------------