当然，这些仍然只是基准测试工具。
问：你们做过性能基准测试么？
此产品的基准测试结果已经可用。
花些时间做些非常简单的基准测试。
要确保基准测试结果是相关的。
首要的问题就是如何设计基准测试。
First and foremost is the question of what should the benchmark look like?
基准测试能够测试许多东西。
图1基准测试的执行栈输出。
基准测试是极其困难的。
这不是一个TPC - H的基准测试结果。
微基准测试应该反映真实用例。
首先进行单节点基准测试，比如STREAM。
本文稍后将会介绍更多有关性能基准测试的内容。
You can read more about performance baselines later in this article.
基准测试是对计算机系统的性能的测试。
A benchmark is a test of the performance of a computer system.
执行许多次基准测试度量。
当然，所有这样的活动都会干扰到我们的基准测试。
Of course all of this activity interferes with our benchmark.
但是，普通的程序员有办法进行正确的基准测试吗？
But is there any hope that you, an ordinary programmer, can do correct benchmarks?
本小节展示了基准测试系统的不同的系统性能测量。
This section presents various system performance measurements of the benchmark system.
这与基准测试完全不同，其中的环境完全受到控制。
This is very different from a benchmark, where the environment is tightly controlled.
我还运行了几个基准测试。
这个基准测试有一个我还没有解释的特性。
There is a peculiarity with this benchmark that I have no explanation of.
课程有项目管理、质控、基准测试等其他。
Courses may include project management, quality control, benchmarking, and many others.
注：本文中的结果都来自于可控环境中执行的基准测试。
Note: The results in this article were from benchmarks executed in a controlled environment.
第一次是不使用电话，也不喝酒的基准测试。
The first was a baseline test with no cellphone usage or alcohol consumption.
非常成功的基准测试系列远不止是性能提升。
The extremely successful benchmark series is much more than a performance boost.
标准偏差总是不可能度量的（微基准测试的典型情况）。
The standard deviations were always impossible to measure (typical of microbenchmarks).
不过，就本文而言，我没有尝试进行任何具体的基准测试。
For this article, however, I have not attempted any specific benchmarks.
由于这些和其他原因，IBM通常依赖行业标准基准测试。
For these and other reasons, IBM generally relies on industry standard benchmarks.
基线测试后，我们立即在同样条件下运行基准测试。
Immediately following the baseline test we run the benchmark under the same conditions.
在相同的物理硬件中，每个基准测试有它自己的LPAR和虚拟化。
Each of these benchmarks had its own LPAR and virtualization within the same physical hardware.
应用推荐