在运行测试前创建统计表。
当有关表的统计信息不准确时，可能会造成性能问题。
When statistical information about tables is not accurate, it may cause performance problems.
最后的统计表是对系统负载情况的度量结果。
The statistics at the end are a measure of how heavily loaded the system is.
表1给出了本文的示例场景收集到的粗略统计。
Table 1 shows the rough statistics gathered for this article's sample scenario.
统计信息显示出表中的数据分布状况。
如果为一个昵称收集了统计信息，那么在数据源上对相应远程表的远程统计信息进行更新时，这些统计信息不会自动更新。
Once the statistics are gathered for a nickname, they are not automatically modified when the remote statistics are updated on the data source for the corresponding remote table.
在缺少当前统计的情况下，优化器选择对参与联接的两个表进行表扫描。
The optimizer, in the absence of current statistics, is choosing to do a table scan on both tables involved in the join.
在研究分布统计信息之前，我们先来看看基本统计信息，只要执行RUNSTATS即可收集这些表的相关统计信息。
Before examining distribution statistics, let's have a look at the basic statistics that are always collected for a table when RUNSTATS is executed.
统计视图是一种强大的、新型的统计，它可以表示复杂谓词或表之间的关系。
Statistical views are a powerful new form of statistics that represent relationships among complex predicates or across tables.
从前面的例子中可以看到，智能统计对于所有表和索引都是通用的，不管是否是片段表或索引。
As you can see from previous example, the smarter statistics is universal to all tables and indexes regardless of whether they are fragmented or not.
RUNSTATS实用程序用于更新系统目录表中的统计信息，以帮助查询优化过程。
The RUNSTATS utility updates statistics in the system catalog tables to help with the query optimization process.
这些统计信息存储在表示声明的临时表的目录信息的内存结构中。
These statistics are stored in memory structures that represent the catalog information for declared temporary tables.
您可以从基础访问路径图中的表节点的表描述符处获得该统计信息。
You can obtain the statistics from the table descriptor of the table node in the underlying access path graph.
而且，现有表没有统计信息。
仅对用户表进行自动统计信息收集。
Only consider user tables for automatic statistics collection.
接下来，收集统计信息，这样系统表中会出现一些数据。
Next, collect statistics so that the system tables will be populated.
清单7给出了与临时表创建有关的统计信息。
Listing 7 shows the statistics associated with temporary-table creation.
例如，可以在片段级别上刷新统计数据，可以向表统计添加或从中移除片段统计。
For example, refreshing of statistics can happen at fragment level, and fragment statistics can be added to, or removed from, table statistics.
数据库服务器不更新系统编目表中的统计信息。
The database server does not update the statistics in the system catalog tables.
对于DETACH操作：生成的表的表级统计是使用片段统计重新构建的。
For DETACH operation: table level statistics of the resulting tables are rebuilt from the fragmented statistics.
仅在收集了一个表中至少一个列的分布统计信息时，SYSCAT.COLDIST 才会包含关于该表的条目。
SYSCAT.COLDIST contains entries for a certain table only if distribution statistics have been collected for at least one of its columns.
您可以使用这个部分创建新表、设置表过滤器或运行更新统计数据。
You can use this section to create new tables, set a filter to tables, or run update statistics.
在我们对基准的第一个改进中，我们向账单表增加索引，并为表更新统计信息。
In our first improvement to the benchmark, let us add an index on the bills table and update statistics for the table.
启动时，动态统计数据集合可通过确保您拥有最新表统计信息来帮助提高数据库性能。
When enabled, automatic statistics collection helps improve database performance by ensuring that you have up-to-date table statistics.
统计数据过时：过时的统计数据无法表示表的当前状态。
Obsolete statistics: Old statistics may no longer represent the current state of the table.
表范围：确定对哪些表执行自动统计信息收集。
Table scope: Identifies which tables are considered for automatic statistics collection.
如果表具有片段级统计，那么将在片段的层面上判断是否执行统计刷新，因此只有片段被修改或出现新片段的情况下重建统计。
If the table has fragment-level statistics, the decision to refresh statistics is done at the fragment level, thereby rebuilding statistics only for changed or new fragments.
因为片段统计比表统计更加细粒度，因此会占用更多的空间。
Since the fragmented statistics are more granular than table statistics, they do take up more space as well.
对于ATTACH操作：将对新的片段构建片段统计，并利用所有的片段统计重新构建表级统计。
For ATTACH operation: fragmented statistics of the new fragment is built, and table level statistics is rebuilt from all fragmented statistics.
相关的输出包括网络连接、路由表、接口统计数据。
Relevant output includes network connections, routing tables, and interface statistics.
应用推荐