correcting range, defaults, and normalizing docs around default_statistics_target (#3819)

78d7d81d · David Yozie · GitHub · bea6a32f · 78d7d81d · 78d7d81d
6 changed file
--- a/gpdb-doc/dita/admin_guide/intro/about_statistics.xml
+++ b/gpdb-doc/dita/admin_guide/intro/about_statistics.xml
@@ -329,8 +329,8 @@
            <codeph>default_statistics_target</codeph> server configuration parameter. The default
          value is usually sufficient, and you should only raise or lower it if your tests
          demonstrate that query plans improve with the new target. For example, to raise the
-          default statistics target from 25 to 50 you can use the <codeph>gpconfig</codeph>
-          utility:<codeblock>gpconfig -c default_statistics_target -v 50</codeblock></p>
+          default statistics target from 100 to 150 you can use the <codeph>gpconfig</codeph>
+          utility:<codeblock>gpconfig -c default_statistics_target -v 150</codeblock></p>
        <p>The statististics target for individual columns can be set with the <codeph>ALTER
            TABLE</codeph> command. For example, some queries can be improved by increasing the
          target for certain columns, especially columns that have irregular distributions. You can

--- a/gpdb-doc/dita/admin_guide/perf_issues.xml
+++ b/gpdb-doc/dita/admin_guide/perf_issues.xml
@@ -135,7 +135,7 @@
          <codeblock>ALTER TABLE sales ALTER COLUMN region SET STATISTICS 50;
 </codeblock>
        </p>
-        <p>This is equivalent to increasing <codeph>default_statistics_target</codeph> for a
+        <p>This is equivalent to changing <codeph>default_statistics_target</codeph> for a
          particular column. Subsequent <codeph>ANALYZE</codeph> operations will then gather more
          statistics data for that column and produce better query plans as a result.</p>
      </body>

--- a/gpdb-doc/dita/best_practices/analyze.xml
+++ b/gpdb-doc/dita/best_practices/analyze.xml
@@ -35,14 +35,14 @@
        quality, or accuracy, of the statistics. </p>
      <p>To allow large tables to be analyzed in a reasonable amount of time,
          <codeph>ANALYZE</codeph> takes a random sample of the table contents, rather than
-        examining every row. To increase sampling for all table columns adjust the
-          <codeph>default_statistics_target</codeph> configuration parameter. The target value
-        ranges from 1 to 1000; the default target value is 25. The
-          <codeph>default_statistics_target</codeph> variable applies to all columns by default. A
-        larger target value increases the time needed to perform the <codeph>ANALYZE</codeph>, but
-        may improve the quality of the query planner’s estimates. This is especially true for
-        columns with irregular data patterns. <codeph>default_statistics_target</codeph> can be set
-        at the master/session level and requires a reload. </p>
+        examining every row. To increase the number of sample values for all table columns adjust
+        the <codeph>default_statistics_target</codeph> configuration parameter. The target value
+        ranges from 1 to 1000; the default target value is 100. The
+          <codeph>default_statistics_target</codeph> variable applies to all columns by default, and
+        specifies the number of values that are stored in the list of common values. A larger target
+        may improve the quality of the query planner’s estimates, especially for columns with
+        irregular data patterns. <codeph>default_statistics_target</codeph> can be set at the
+        master/session level and requires a reload. </p>
      <p>The <codeph>gp_analyze_relative_error</codeph> configuration parameter affects the sampling
        rate during statistics collection to determine cardinality in a column. For example, a value
        of .5 is equivalent to an acceptable error of 50%. The default is .25. Use the

--- a/gpdb-doc/dita/ref_guide/config_params/guc-list.xml
+++ b/gpdb-doc/dita/ref_guide/config_params/guc-list.xml
@@ -1659,10 +1659,10 @@
  <topic id="default_statistics_target">
    <title>default_statistics_target</title>
    <body>
-      <p>Sets the default statistics target for table columns that have not had a column-specific
-        target set via <codeph>ALTER TABLE SET STATISTICS</codeph>. Larger values increase the time
-        needed to do <codeph>ANALYZE</codeph>, but may improve the quality of the legacy query
-        optimizer (planner) estimates.</p>
+      <p>Sets the default statistics sampling target (the number of values that are stored in the
+        list of common values) for table columns that have not had a column-specific target set via
+          <codeph>ALTER TABLE SET STATISTICS</codeph>. Larger values may improve the quality of the
+        legacy query optimizer (planner) estimates. </p>
      <table id="default_statistics_target_table">
        <tgroup cols="3">
          <colspec colnum="1" colname="col1" colwidth="1*"/>

--- a/gpdb-doc/dita/ref_guide/sql_commands/ALTER_TABLE.xml
+++ b/gpdb-doc/dita/ref_guide/sql_commands/ALTER_TABLE.xml
@@ -150,7 +150,7 @@ ALTER TABLE <varname>name</varname>
                                <li id="ay136944"><b>SET STATISTICS</b> — Sets the per-column
                                        statistics-gathering target for subsequent
                                                <codeph>ANALYZE</codeph> operations. The target can
-                                        be set in the range 0 to 1000, or set to -1 to revert to
+                                        be set in the range 1 to 1000, or set to -1 to revert to
                                        using the system default statistics target
                                                (<codeph>default_statistics_target</codeph>).</li>
                                <li id="ay137022"><b>ADD <varname>table_constraint</varname></b> —

--- a/gpdb-doc/dita/ref_guide/sql_commands/ANALYZE.xml
+++ b/gpdb-doc/dita/ref_guide/sql_commands/ANALYZE.xml
@@ -157,7 +157,7 @@
        column-by-column basis by setting the per-column statistics target with <codeph>ALTER TABLE
          ... ALTER COLUMN ... SET STATISTICS</codeph> (see <codeph>ALTER TABLE</codeph>). The
        target value sets the maximum number of entries in the most-common-value list and the
-        maximum number of bins in the histogram. The default target value is 10, but this can be
+        maximum number of bins in the histogram. The default target value is 100, but this can be
        adjusted up or down to trade off accuracy of planner estimates against the time taken for
          <codeph>ANALYZE</codeph> and the amount of space occupied in
        <codeph>pg_statistic</codeph>. In particular, setting the statistics target to zero disables