MMM - Matrix-Matrix Multiplicaion Benchmark | |||||
| 处理器 | 双路Intel Nehalem-EP Xeon X5570 | 双路Intel Westmere-EP Xeon X5670 | 双路Intel Westmere-EP Xeon X5680 | DELL PowerEdge M910 四路Intel Nehalem-EX Xeon E7540 | 四路Intel Nehalem-EX Xeon X7560 |
| 单位 | GFLOPS | GFLOPS | GFLOPS | GFLOPS | GFLOPS |
Threads 1 | |||||
| 5000 step | 7.821975 | 7.842319 | 8.877563 | 5.867645 | 7.036748 |
| 10000 step | 7.890761 | 7.840417 | 8.883291 | 5.865347 | 7.034792 |
| 15000 step | 7.888751 | 7.845479 | 8.881528 | 5.826604 | 6.960592 |
Threads 2 | |||||
| 5000 step | 15.59136 | 15.62796 | 17.5891 | 11.570261 | 13.979099 |
| 10000 step | 15.7544 | 15.66469 | 17.73566 | 11.689317 | 14.032266 |
| 15000 step | 15.7445 | 15.64657 | 17.67208 | 11.602247 | 13.89951 |
Threads 4 | |||||
| 5000 step | 30.69218 | 29.99696 | 34.85343 | 21.788607 | 27.555005 |
| 10000 step | 31.02227 | 29.75883 | 34.90105 | 22.21115 | 27.685804 |
| 15000 step | 31.04954 | 30.55926 | 34.92557 | 22.073702 | 27.297404 |
Threads 8 | |||||
| 5000 step | 36.2252 | 49.03697 | 45.99856 | 41.228878 | 49.518835 |
| 10000 step | 38.21083 | 50.30305 | 45.99856 | 43.472432 | 49.767277 |
| 15000 step | 40.71236 | 56.00031 | 47.74417 | 43.37777 | 49.750117 |
Threads 16 | |||||
| 5000 step | 59.38371 | 64.04222 | 66.10022 | 73.379889 | 47.855051 |
| 10000 step | 61.44583 | 62.42291 | 72.38159 | 78.596851 | 48.320744 |
| 15000 step | 61.83442 | 64.3761 | 73.2495 | 79.099092 | 48.421492 |
Threads 24 | |||||
| 5000 step | 54.82514 | 84.13599 | 66.10022 | 94.000418 | 70.233111 |
| 10000 step | 54.82514 | 88.58685 | 72.38159 | 124.028823 | 71.502532 |
| 15000 step | 59.18915 | 90.12297 | 73.2495 | 124.574801 | 71.685326 |
Threads 32 | |||||
| 5000 step | 96.076302 | ||||
| 10000 step | 97.64478 | ||||
| 15000 step | 98.195937 | ||||
Threads 48 | |||||
| 5000 step | 97.335138 | 136.623189 | |||
| 10000 step | 119.780984 | 142.50301 | |||
| 15000 step | 121.637469 | 145.534244 | |||
Threads 64 | |||||
| 5000 step | 137.141474 | ||||
| 10000 step | 182.49902 | ||||
| 15000 step | 185.098571 | ||||
MMM是一个类似矩阵乘法基准测试软件，得到的结果单位是GFLOPS，也就是说它是一个浮点测试。可以看到，X7560平台最终展示了185.1GFLOPS的计算能力。MMM和Linpack一样，都能充分地利用CPU运算核心的能力，因此它实际上建议关闭超线程来测试。
SunGard Adaptiv Analytics Benchmark v4.0 | |||||
| 处理器 | 双路Intel Nehalem-EP Xeon X5570 | 双路Intel Westmere-EP Xeon X5670 | 双路Intel Westmere-EP Xeon X5680 | DELL PowerEdge M910 四路Intel Nehalem-EX Xeon E7540 | 四路Intel Nehalem-EX Xeon X7560 |
| Threads | 16 | 24 | 24 | 48 | 64 |
| Time (lower is better) | 138.076s | 110.331s | 94.911s | 139.512s | 104.925s |
对于SunGard风险分析管理套件基准测试程序来说，多线程有其优势，不过高频率带来的好处也不少，X7560平台的表现不错。
black_scholes | |||||
| 处理器 | 双路Intel Nehalem-EP Xeon X5570 | 双路Intel Westmere-EP Xeon X5670 | 双路Intel Westmere-EP Xeon X5680 | DELL PowerEdge M910 四路Intel Nehalem-EX Xeon E7540 | 四路Intel Nehalem-EX Xeon X7560 |
| Threads | 16 | 24 | 24 | 48 | 64 |
| Time (lower is better) | 9.17s | 6.16s | 5.51s | 4.40s | 2.78s |
black_scholes是对布莱克-肖尔斯期权定价模型进行计算，布莱克-肖尔斯期权定价模型是由1997诺贝尔经济学奖的两个获得者创立和发展的模型。看起来这个测试对多线程的支持不错，X7560平台成绩是2.78秒。
