SiSoftware Sandra是一款可运行在32bit和64bit Windows操作系统上的分析软件，这款软件可以对于系统进行方便、快捷的基准测试，还可以用于查看系统的软件、硬件等信息。从Sandra 2007开始支持SSE4指令集。SiSoftware Sandra所有的基准测试都针对SMP和SMT进行了优化，最高可支持32/64路平台，这也是我们选择这款软件的原因之一。
SiSoftware Sandra Pro Business 2009 | |||
测试对象 | DELL Precision T5500 | HP xw8600 Workstation | |
Processor Arithmetic Benchmark | |||
Dhrystone ALU | 138679MIPS | 98048MIPS | |
Dhrystone ALU vs SPEED | 52.13MIPS/MHz | 36.76MIPS/MHz | |
Whetstone iSSE3 | 121017MFLOPS | 72625MFLOPS | |
Dhrystone iSSE3 vs SPEED | 45.50MFLOPS/MHz | 27.23MFLOPS/MHz | |
Processor Multi-Media Benchmark | |||
Multi-Media Int x8 iSSE4.1 | 284.58MPixel/s | 253.01MPixel/s | |
Multi-Media Int x8 iSSE4.1 vs SPEED | 106.99kPixels/s/MHz | 94.87kPixels/s/MHz | |
Multi-Media Float x4 iSSE2 | 217.19MPixel/s | 169.32MPixel/s | |
Multi-Media Float x4 iSSE2 vs SPEED | 81.65kPixels/s/MHz | 63.49kPixels/s/MHz | |
Multi-Media Double x2 iSSE2 | 119.23MPixel/s | 89.47MPixel/s | |
Multi-Media Double x2 iSSE2 vs SPEED | 44.82kPixels/s/MHz | 33.55kPixels/s/MHz | |
Multi-Core Efficiency Benchmark | |||
Inter-Core Bandwidth | 69.66GB/s | 22.42GB/s | |
Inter-Core Bandwidth vs SPEED | 26.82MB/s/MHz | 8.61MB/s/MHz | |
Inter-Core Latency | 17ns | 90ns | |
Inter-Core Latency vs SPEED | 0.01ns/MHz | 0.03ns/MHz | |
Memory Bandwidth Benchmark | |||
Int Buff'd iSSE2 Memory Bandwidth | 15.08GB/s | 4.14GB/s | |
Int Buff'd iSSE2 Memory Bandwidth vs SPEED | - | 6.36MB/s/MHz | |
Float Buff'd iSSE2 Memory Bandwidth | 15.08GB/s | 4.14GB/s | |
Float Buff'd iSSE2 Memory Bandwidth vs SPEED | - | 6.36MB/s/MHz | |
Memory Latency Benchmark | |||
Memory(Random Access) Latency | 114ns | 109ns | |
Memory(Random Access) Latency vs SPEED | - | 0.16ns/MHz | |
Speed Factor | 78.00 | 95.50 | |
Internal Data Cache Latency | 4clocks | 3clocks | |
L2 On-board Cache Latency | 10clocks | 18clocks | |
Cache and Memory Benchmark | |||
Cache/Memory Bandwidth | 128.85GB/s | 79.01GB/s | |
Cache/Memory Bandwidth vs SPEED | 49.60MB/s/MHz | 30.34MB/s/MHz | |
Speed Factor | 25.70 | 86.60 | |
Internal Data Cache | 424.43GB/s | 419.51GB/s | |
L2 On-board Cache | 395.93GB/s | 140.91GB/s | |
Graphics Processing | |||
Native Float Shaders | 537.10MPixel/s | 133.71MPixel/s | |
Native Float Shaders vs SPEED | 542.11kPixels/s/MHz | 123.58kPixels/s/MHz | |
Emulated Double Shaders | - | 12.68MPixel/s | |
Emulated Double Shaders vs SPEED | - | 11.71kPixels/s/MHz | |
Graphics Bandwidth | |||
Internal Memory Bandwidth | 78.03GB/s | 45.44GB/s | |
Internal Memory Bandwidth vs SPEED | - | 66.28MB/s/MHz | |
Data Transfer Bandwidth | 9.25GB/s | 1.97GB/s | |
Data Transfer Bandwidth vs SPEED | - | 2.88MB/s/MHz | |
Internal Memory Bandwidth | 78.03GB/s | 45.44GB/s | |
Internal Memory Bandwidth Efficiency | - | 69.04% | |
System to Device Bandwidth | 9.58GB/s | 2.00GB/s | |
System to Device Bandwidth Efficiency | 61.30% | 51.07% | |
Device to System Bandwidth | 6.34GB/s | 1.76GB/s | |
Device to System Bandwidth Efficiency | 40.57% | 45.15% | |
受益于Nehalem-EP处理器的优势，Precision T5500在大多数项目中的成绩都领先于对比平台，在处理器和内存的相关测试中表现得更为明显。由于处理器的差异，两者测试的项目并不完全对等，有些项目Nehalem-EP没有相关的测试成绩。不过由此看出，在使用QPI架构、内置三通道DDR3内存控制器之后，新一代的Precision T5500工作站相比老款的至强5400系列来说有了明显的提升，个别项目性能翻倍。
