问题在于统计数据并不是对现实的客观衡量；它们只是一个最佳估测值。
The problem is that the statistics aren't an objective measure of reality; they are simply a best approximation.
跟踪连接或语句的元数据信息让我可以跨越请求，针对连接或语句把统计值分组。
Tracking metadata information about connections or statements lets me group statistics across each request for equivalent connections or statements.
在我们运行过程中，我们看到你可以让所有的统计值是正确的，但依然得到了错误的答案,这可能是有什么连续性错误。
One of the things we'll see as we go on, is you can get all your statistics right, and still get the wrong answer because of a consistent bug.
在第二部分，我将重新回到这个主题，介绍如何更新J MX支持来显示像这样的嵌套统计值。
In part two, I'll revisit this topic to show how I update the JMX support to display nested statistics like this.
这些统计值容易搜集，但是在需要数据的时候，如果没有非侵入性的数据检索机制，那么这些值就不太有用。
These statistics are easy to gather, but without an unintrusive means of retrieving the data when it is needed, they are not very useful.
这个方面有两个主要职责：跟踪与创建和准备语句有关的信息，然后监视jdbc语句执行的性能统计值。
This aspect has two primary responsibilities: tracking information about creating and preparing statements and then monitoring performance statistics for executing JDBC statements.
比起在整个示例应用程序中用分散的调用更新统计值和跟踪上下文，这是一个重大的改进。
This is a major improvement over scattering calls to update statistics and to track context throughout my example application.
它调用了函数，这个函数通过，调用这个函数执行一定数量的实验，然后我们计算这些统计值并将其表示出来。
It calls this function, which runs an appropriate number of trials by calling that function, and we'll calculate and present some statistic.
建议的先后次序在这里没影响；虽然是在执行完成之后才用建议查询统计值，但字符串是在执行发生之前捕捉的。
Advice precedence doesn't matter here; the string is captured before the execution occurs, whereas it's used to look up statistics after execution has completed.
您可能对更改其中某些统计值来影响优化器、或者在开发或测试环境中调查研究数据库性能感兴趣。
You might be interested in changing some of these statistical values to influence the optimizer or to investigate database performance in a development or test environment.
每个基类负责跟踪某项操作前后的性能，还需要更新系统范围内这条信息的性能统计值。
Each base class is responsible for tracking performance before and after certain operations and will need to update system-wide performance statistics for that information.
自然也可以把这种方式扩展成计算额外的统计值，并在问题可能发生的地方保存单独的数据点。
Naturally, you could extend this approach to calculate additional statistics and to store individual data points where issues might arise.
有关计算t统计值概率的更多信息，请参阅第1部分。
For more on computing the probability of the t statistic, see Part 1.
那么，如何计算t统计值的概率呢？
随后在执行jdbc语句更新对应请求的统计值时，会使用这个图。
This map is used subsequently when JDBC statements are executed to update statistics for the appropriate request.
一个重要的汇总值是T统计值，它可以用来衡量一个线性方程与数据的吻合程度。
One important summary value is a t statistic that can be used to measure how well a linear equation fits the data.
它允许捕捉请求的总数、总时间以及最差情况性能之类的统计值，还允许深入请求中数据库调用的信息。
It lets you capture statistics such as total counts, total time, and worst-case performance for requests, and it will also let you drill down into that information for database calls within a request.
欧洲的科学家找来了一些志愿者，叫他们衡量一些统计值，比如瑞士的人口密度等等。
European scientists asked volunteers to estimate statistics like the population density of Switzerland.
这个过程基于对t统计值的计算，使用概率函数求得随机大的观测值的概率。
This procedure is based upon computing a t statistic and using a probability function to find the probability of observing a value that large by chance.
计算t统计值概率。
例如，我可以用带有更底层机制的单体(例如ThreadLocal)持有一堆统计值和上下文。
For example, I could have used a singleton with lower-level mechanisms like a ThreadLocal holding a stack of statistics and context.
PerformGC按钮则证明了JMX可以提供除了查看操作统计值之外的初始化操作的功能。
The Perform GC button illustrates that JMX offers the capability to initiate operations in addition to viewing operating statistics.
jstat——得到进程id的统计值(jstat - gcpid)。
low模式只填充表的标量统计值(也就是说，没有分布信息)。
The low mode populates only the scalar statistical values (that is, no distribution information) of a table.
其中包括计数、总时间、最大时间的统计值，还有重设操作，这个接口是PerfStats接口的子集。
This consists of the statistics values for counts, total time, maximum time, and the reset operation, which is a subset of the PerfStats interface.
分位数统计信息提供关于一个列中的值是否聚合的信息。
Quantile statistics provide the information whether the values of a column are clustered or not.
虽然这个访问计划比使用具体值但没有分布统计信息情况下的访问计划好，但不如既使用具体值又有分布统计信息时的访问计划。
Although the access plan is better than the access plan with concrete values without distribution statistics, it is worse than the access plan with concrete values with distribution statistics.
虽然这个访问计划比使用具体值但没有分布统计信息情况下的访问计划好，但不如既使用具体值又有分布统计信息时的访问计划。
Although the access plan is better than the access plan with concrete values without distribution statistics, it is worse than the access plan with concrete values with distribution statistics.
应用推荐