现在让基准运行时间更长一些，来再次进行检查。
Let's let the benchmark run a little longer and check again.
延迟测试的基准运行(使用WebSphereMQLowLatencyMessaging的默认设置)的结果如图5所示。
A baseline run of the latency test, using default Settings for WebSphere MQ Low latency Messaging, gives the results shown in Figure 5.
又或者，必须运行性能测试来确定初始基准。
Or, performance tests have to be run to determine initial benchmarks.
在运行基准测试或其他性能测试时，必须谨慎地控制系统上的其他工作。
When running the baseline test or other performance tests, other work on the system must be carefully controlled.
您最好是在一系列不同的平台上运行一系列的基准，然后寻找结果中的相似之处。
The best you can do is run a series of benchmarks on a series of different platforms, and look for similarity in the results.
代码达到稳定状态之后，基准必须对这段代码运行多次，然后才能对结果做出有效的统计分析。
Once the code is in a steady state then the benchmark must run it several times and compute a statistical analysis of the results.
虽然关于如何运行基准的细节超出了我们的讨论范围，但仍有必要介绍一遍一些关键的方面。
Although the details of how to run the benchmark is out of the scope of our discussion, it is still worth going through some of the key aspects.
由于用户增加，这就需要运行基准测试来模拟新增的用户。
With these "bigger Numbers" comes the requirement of running benchmarks to simulate these increased user levels.
这可以通过运行几个行业认可的基准测试来完成。
This can be accomplished by running a few industry-accepted benchmarks.
在内部我们运行了大量的基准，当然，有一些结果是针对6.4的，但是我们没时间发布他们。
We run a number of benchmarks internally, so yes, there are some benchmark results for 6.4, but we have had no time to publish them yet.
所以，针对验证目的而运行基准检验时，必须确保验证期间不会发生数据刷新。
So, when running a benchmark for validation purposes you must ensure that your validation happens during a time when the data will not refresh.
要帮助回答这个问题，您必须主动地收集并且维护一个基准：有关系统正常运行时的状态的广泛信息。
To help answer this question, you must actively collect and maintain a baseline: extensive information about the state of the system at a time when that system is operating normally.
表 1 到表4 所示的用户数只是运行基准脚本的实例，它们不一定与服务器上部署的实际用户数相关。
The users shown in tables 1-4 are only instances of the benchmark script running, and they do not necessarily correlate to the number of actual users deployed on a server.
在基准测试环境中，您可以控制安装，正确地设置哪些程序运行在服务器上，从基准测试客户端以某个特定的比率。
In a benchmark environment, you have control over the setup, and you establish exactly what is running on your server, at what specific rates, from your benchmark clients.
使用-基准，看看哪种方法会在你的机器上最快的运行。
Use — benchmark to see which method does perform best on your machine.
如果某个计算节点崩溃，管理节点可以把它记录下来，您将知道尽管运行的基准不正确，但是已经执行了适当的动作来恢复出现故障的计算节点。
If a compute node crashes, the management node can log it and you will know that while the benchmark that ran is not accurate, the proper action is being taken to restore the failed compute node.
这使得您能够在系统正常运行时建立基准数据，以便可以使用本系列文章中所介绍的方法，包括优化您的内存子系统。
This enables you to establish a baseline while your system is healthy so that you can practice some of the methods discussed in this series, which include tuning your memory subsystems.
基于本文的目的，我想说，基准的运行时间过长，我想在一个小时内执行完毕。
For the purposes of this article, I will contend that the benchmark takes too long to run and that we need it to execute in under an hour.
我还简要地介绍了优化方法和建立系统正常运行时的基准数据的重要性。
I also briefly discussed tuning methodology and the importance of establishing a baseline while the system is behaving normally.
未修改过的基准代码的运行时间大约为30分钟，要求参赛者改进它，让它运行得更快。
The benchmark unchanged took about 30 minutes to run and I challenged participants to make it run faster.
我们将使用Ganglia来检验运行Linpack基准的属性。
We'll use Ganglia to examine the attributes of running the Linpack benchmark.
事实上，WebSpherePerformanceLab在运行涉及Trade性能基准时，使用了12个CPU(有时更多)。
In fact, the WebSphere performance Lab has fully utilized 12 CPUs (and in some cases more) when running tests involving the Trade performance benchmark.
基准测试机器运行32位版本的Informix并拥有2GB的内存限制(64位版本的Informix没有此限制)。
The benchmark machine was running a 32-bit version of Informix and was limited to 2 GB of memory (the 64-bit version of Informix does not have this limit).
在Core2Duo上，第二个核将固定的运行基准从3731ms提高到了6574ms，或者说增加了176%。
With the Core 2 Duo, that second core increased the runtime of the fixed work benchmark from 3731 ms. to 6574 ms. or a factor of 176%.
现在就建立起了一个小型的高性能集群，可以试运行并行构建的Linpack基准。
You should now have a small high performance cluster set up that can exercise the parallel built Linpack benchmark.
为了回答这些问题，我重新运行这些基准测试，但是这一次随机选择integers的元素，而不是顺序选择。
To answer these questions, I reran those benchmarks but picked random elements of integers instead of sequential elements.
最好的选择是，在BIOS中关闭Core2Duo中的一个核，然后重新运行基准测试。
The best option turned out to be re-running the benchmark on my machine with one of the cores in the Core 2 Duo turned off, set in the BIOS.
测试脚本：下面是测试使用的脚本(关于运行Sort基准的更多信息参见HadoopWiki)。
Testing scripts: Following are the scripts used for testing (refer to the Hadoop Wiki for more information about running the Sort benchmark).
即使您编写了一个很好的基准，得到的结果可能也只是在运行基准的系统上才有效。
Even when you write an excellent benchmark, your results may be only valid on the system you ran it on.
除了评测使用的资源外，我们还观察了运行基准测试用户的成本。
In addition to evaluating the resources used, we also looked at the cost of running the benchmark users.
应用推荐