------------------------------------------------------------- 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 8 174.5506 0.4584 0.4583 0.4585 read 32 213.9800 0.3739 0.3739 0.3740 read 64 214.9746 0.3723 0.3721 0.3725 read 32x2 214.1007 0.3737 0.3737 0.3738 read 32 CP3 268.7035 0.2980 0.2977 0.2982 read 32 CP4 268.2350 0.2983 0.2982 0.2983 read 32 CP5 268.4402 0.2981 0.2980 0.2981 read 32 CP6 267.0083 0.2999 0.2996 0.3002 read 32x4 CP3 268.4492 0.2988 0.2980 0.2995 read 32x4 CP4 267.1448 0.2995 0.2995 0.2996 read 32x4 CP5 267.5022 0.2991 0.2991 0.2991 read 32x4 CP6 266.7761 0.3000 0.2999 0.3001 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 248.4433 0.3224 0.3220 0.3228 write 32 275.4212 0.2907 0.2905 0.2909 write 64 615.8494 0.1299 0.1299 0.1300 write 32x2 278.2213 0.2897 0.2875 0.2918 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 192.9931 0.8293 0.8290 0.8297 cmp 32 223.3620 0.7257 0.7163 0.7350 cmp 64 227.3447 0.7042 0.7038 0.7046 cmp 32x2 223.7149 0.7157 0.7152 0.7162 cmp 32 CP2 267.4525 0.5988 0.5982 0.5994 cmp 32 CP3 268.8141 0.5957 0.5952 0.5962 cmp 32 CP4 267.3747 0.5985 0.5984 0.5985 cmp 32 CP5 267.3926 0.5986 0.5984 0.5989 cmp 32 CP6 268.0610 0.5973 0.5969 0.5977 cmp 32x4 CP2 267.5362 0.5984 0.5980 0.5988 cmp 32x4 CP3 269.0015 0.5953 0.5948 0.5958 cmp 32x4 CP4 266.2397 0.6012 0.6010 0.6014 cmp 32x4 CP5 267.8940 0.5977 0.5973 0.5981 cmp 32x4 CP6 265.7635 0.6021 0.6020 0.6022 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 160.4779 0.9982 0.9970 0.9993 copy 32 155.3301 1.0334 1.0301 1.0367 copy 64 152.3254 1.0518 1.0504 1.0531 copy 32x2 177.3958 0.9020 0.9019 0.9020 copy 32 CP2 209.4935 0.7664 0.7637 0.7690 copy 32 CP3 244.6087 0.6570 0.6541 0.6598 copy 32 CP4 252.7610 0.6361 0.6330 0.6392 copy 32 CP5 253.0644 0.6352 0.6323 0.6382 copy 32x4 CP2 239.7435 0.6699 0.6674 0.6725 copy 32x4 CP3 256.0562 0.6287 0.6249 0.6325 copy 32x4 CP4 263.2917 0.6103 0.6077 0.6130 copy 32x4 CP5 263.6379 0.6111 0.6069 0.6152 copy 64x4 CP4 276.4349 0.5816 0.5788 0.5844 glibcb memcpy 178.1908 0.8992 0.8979 0.9005 bmove512 155.6844 1.0281 1.0277 1.0286 FC64 268.2111 0.5999 0.5965 0.6032 ------------------------------------------------------------- 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 8 357.4587 0.2240 0.2238 0.2241 read 32 1411.6057 0.0568 0.0567 0.0569 read 64 1272.7078 0.0630 0.0629 0.0632 read 32x2 1500.6253 0.0534 0.0533 0.0535 read 32 CP3 1200.9288 0.0668 0.0666 0.0671 read 32 CP4 1212.4719 0.0663 0.0660 0.0665 read 32 CP5 1207.1286 0.0665 0.0663 0.0668 read 32 CP6 1209.7224 0.0663 0.0661 0.0664 read 32x4 CP3 1812.7047 0.0445 0.0441 0.0448 read 32x4 CP4 1821.0768 0.0451 0.0439 0.0463 read 32x4 CP5 1818.3038 0.0442 0.0440 0.0444 read 32x4 CP6 1813.1552 0.0443 0.0441 0.0445 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 537.6891 0.1488 0.1488 0.1489 write 32 1343.0261 0.0598 0.0596 0.0600 write 64 1675.8530 0.0478 0.0477 0.0479 write 32x2 1342.0807 0.0598 0.0596 0.0600 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 439.0818 0.3650 0.3644 0.3656 cmp 32 992.3765 0.1614 0.1612 0.1616 cmp 64 1058.7549 0.1516 0.1511 0.1521 cmp 32x2 995.2419 0.1609 0.1608 0.1610 cmp 32 CP2 962.6325 0.1675 0.1662 0.1688 cmp 32 CP3 1014.7653 0.1578 0.1577 0.1579 cmp 32 CP4 1062.8969 0.1508 0.1505 0.1510 cmp 32 CP5 1085.0471 0.1484 0.1475 0.1493 cmp 32 CP6 1106.9302 0.1450 0.1445 0.1455 cmp 32x4 CP2 1208.0499 0.1392 0.1324 0.1459 cmp 32x4 CP3 1287.9865 0.1242 0.1242 0.1242 cmp 32x4 CP4 1306.8352 0.1236 0.1224 0.1247 cmp 32x4 CP5 1364.9196 0.1173 0.1172 0.1173 cmp 32x4 CP6 1378.3224 0.1172 0.1161 0.1182 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 463.3233 0.3464 0.3453 0.3476 copy 32 1129.7218 0.1520 0.1416 0.1623 copy 64 2536.3724 0.0632 0.0631 0.0633 copy 32x2 1143.6334 0.1406 0.1399 0.1413 copy 32 CP2 1054.8806 0.1521 0.1517 0.1526 copy 32 CP3 1058.6347 0.1519 0.1511 0.1526 copy 32 CP4 1093.7478 0.1463 0.1463 0.1464 copy 32 CP5 1097.8740 0.1470 0.1457 0.1483 copy 32x4 CP2 1205.3637 0.1328 0.1327 0.1329 copy 32x4 CP3 1221.8785 0.1315 0.1309 0.1321 copy 32x4 CP4 1235.1219 0.1303 0.1295 0.1311 copy 32x4 CP5 1236.7516 0.1294 0.1294 0.1295 copy 64x4 CP4 2816.8005 0.0569 0.0568 0.0570 glibcb memcpy 1144.6731 0.1406 0.1398 0.1414 bmove512 1175.3730 0.1365 0.1361 0.1368 FC64 1745.9807 0.0931 0.0916 0.0945 ------------------------------------------------------------- 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 8 393.0798 0.2037 0.2035 0.2039 read 32 2437.0788 0.0328 0.0328 0.0328 read 64 1238.9847 0.0650 0.0646 0.0654 read 32x2 2646.8122 0.0307 0.0302 0.0311 read 32 CP3 1555.7219 0.0516 0.0514 0.0517 read 32 CP4 1554.0647 0.0515 0.0515 0.0516 read 32 CP5 1561.2159 0.0513 0.0512 0.0514 read 32 CP6 1560.3084 0.0514 0.0513 0.0516 read 32x4 CP3 2474.1142 0.0324 0.0323 0.0324 read 32x4 CP4 2484.2438 0.0322 0.0322 0.0323 read 32x4 CP5 2463.3434 0.0326 0.0325 0.0327 read 32x4 CP6 2475.7754 0.0323 0.0323 0.0323 ------------------------------------------------------------- Write test (setting array A). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time write 8 675.4643 0.1186 0.1184 0.1188 write 32 2592.3570 0.0312 0.0309 0.0315 write 64 2098.6998 0.0381 0.0381 0.0381 write 32x2 2652.9647 0.0302 0.0302 0.0303 ------------------------------------------------------------- Compare test (comparing the source and destination arrays). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time cmp 8 549.0223 0.2924 0.2914 0.2935 cmp 32 2543.6789 0.0630 0.0629 0.0631 cmp 64 2044.3316 0.0785 0.0783 0.0787 cmp 32x2 2608.4588 0.0613 0.0613 0.0614 cmp 32 CP2 1573.8735 0.1018 0.1017 0.1020 cmp 32 CP3 1574.3387 0.1025 0.1016 0.1035 cmp 32 CP4 1572.6969 0.1019 0.1017 0.1020 cmp 32 CP5 1575.5622 0.1016 0.1016 0.1016 cmp 32 CP6 1564.4988 0.1023 0.1023 0.1023 cmp 32x4 CP2 2274.7302 0.0705 0.0703 0.0707 cmp 32x4 CP3 2274.5375 0.0704 0.0703 0.0705 cmp 32x4 CP4 2274.7302 0.0705 0.0703 0.0707 cmp 32x4 CP5 2274.7996 0.0711 0.0703 0.0719 cmp 32x4 CP6 2257.2093 0.0710 0.0709 0.0710 ------------------------------------------------------------- Copy test (copying array A -> B). ------------------------------------------------------------- Function Rate (MB/s) Avg time Min time Max time copy 8 528.6864 0.3029 0.3026 0.3031 copy 32 2112.6270 0.0758 0.0757 0.0759 copy 64 4093.6513 0.0397 0.0391 0.0403 copy 32x2 2120.1557 0.0757 0.0755 0.0758 copy 32 CP2 1279.2436 0.1253 0.1251 0.1255 copy 32 CP3 1276.0984 0.1254 0.1254 0.1255 copy 32 CP4 1280.9113 0.1249 0.1249 0.1250 copy 32 CP5 1279.3460 0.1259 0.1251 0.1268 copy 32x4 CP2 2523.3070 0.0635 0.0634 0.0636 copy 32x4 CP3 2523.4588 0.0637 0.0634 0.0640 copy 32x4 CP4 2524.2941 0.0635 0.0634 0.0636 copy 32x4 CP5 2507.5989 0.0643 0.0638 0.0648 copy 64x4 CP4 4102.6859 0.0390 0.0390 0.0391 glibcb memcpy 1997.2044 0.0802 0.0801 0.0803 bmove512 1638.0681 0.0981 0.0977 0.0985 FC64 3712.3689 0.0433 0.0431 0.0434 -------------------------------------------------------------