Additional docs for toast_tuple_target changes
This commit is contained in:
parent
59af8d4384
commit
117469006b
|
@ -429,7 +429,7 @@ when a row value to be stored in a table is wider than
|
|||
<symbol>TOAST_TUPLE_THRESHOLD</symbol> bytes (normally 2 kB).
|
||||
The <acronym>TOAST</acronym> code will compress and/or move
|
||||
field values out-of-line until the row value is shorter than
|
||||
<symbol>TOAST_TUPLE_TARGET</symbol> bytes (also normally 2 kB)
|
||||
<symbol>TOAST_TUPLE_TARGET</symbol> bytes (also normally 2 kB, adjustable)
|
||||
or no more gains can be had. During an UPDATE
|
||||
operation, values of unchanged fields are normally preserved as-is; so an
|
||||
UPDATE of a row with out-of-line values incurs no <acronym>TOAST</acronym> costs if
|
||||
|
@ -483,6 +483,11 @@ of that data type, but the strategy for a given table column can be altered
|
|||
with <link linkend="sql-altertable"><command>ALTER TABLE ... SET STORAGE</command></link>.
|
||||
</para>
|
||||
|
||||
<para>
|
||||
<symbol>TOAST_TUPLE_TARGET</symbol> can be adjusted for each table using
|
||||
<link linkend="sql-altertable"><command>ALTER TABLE ... SET (toast_tuple_target = N)</command></link>
|
||||
</para>
|
||||
|
||||
<para>
|
||||
This scheme has a number of advantages compared to a more straightforward
|
||||
approach such as allowing row values to span pages. Assuming that queries are
|
||||
|
|
Loading…
Reference in New Issue