manual.texi Fix a few agreement problems.

parent 3aa161d2
...@@ -54017,7 +54017,7 @@ T ...@@ -54017,7 +54017,7 @@ T
@item @item
Fixed optimiser problem on @code{SELECT} when using many overlapping indexes. Fixed optimiser problem on @code{SELECT} when using many overlapping indexes.
MySQL should now be able to choose keys even better when there MySQL should now be able to choose keys even better when there
is many keys to choose from. are many keys to choose from.
@item @item
Changed optimiser to prefer a range key instead of a ref key when the Changed optimiser to prefer a range key instead of a ref key when the
range key can uses more columns than the ref key (which only can use range key can uses more columns than the ref key (which only can use
...@@ -58264,25 +58264,25 @@ Instead of using row level locks one can use application level locks ...@@ -58264,25 +58264,25 @@ Instead of using row level locks one can use application level locks
only in well-behaved applications. only in well-behaved applications.
In many cases one can do an educated guess which locking type is best In many cases one can do an educated guess which locking type is best
for the application but generally it's very hard to say that a given for the application, but generally it's very hard to say that a given
lock type is better than another; everything depends on the application lock type is better than another; everything depends on the application
and different part of the application may require different lock types. and different part of the application may require different lock types.
Here are some tips about locking in MySQL: Here are some tips about locking in MySQL:
On web application most applications do lots of selects, very few Most web applications do lots of selects, very few
deletes, updates mainly on keys and inserts in some specific tables. deletes, updates mainly on keys, and inserts in some specific tables.
The base MySQL setup is very well tuned for this. The base MySQL setup is very well tuned for this.
Concurrent users is not a problem if one doesn't mix updates and selects Concurrent users are not a problem if one doesn't mix updates with selects
that needs to examine many rows in the same table. that need to examine many rows in the same table.
If one mixes inserts and deletes on the same table then @code{INSERT DELAYED} If one mixes inserts and deletes on the same table then @code{INSERT DELAYED}
may be of great help. may be of great help.
One can also use @code{LOCK TABLES} to speed up things (many updates within One can also use @code{LOCK TABLES} to speed up things (many updates within
a single lock is much faster than updates without locks). Splitting a single lock is much faster than updates without locks). Splitting
thing to different tables will also helps. thing to different tables will also help.
If you get speed problems with the table locks in MySQL, you If you get speed problems with the table locks in MySQL, you
may be able to solve these by converting some of your tables to @code{InnoDB} may be able to solve these by converting some of your tables to @code{InnoDB}
...@@ -58290,7 +58290,7 @@ or @code{BDB} tables. ...@@ -58290,7 +58290,7 @@ or @code{BDB} tables.
@xref{InnoDB}. @xref{BDB}. @xref{InnoDB}. @xref{BDB}.
The optimisation section in the manual covers a lot of different aspects of The optimisation section in the manual covers a lot of different aspects of
how to tune ones application. @xref{Tips}. how to tune applications. @xref{Tips}.
@node RTS-threads, Thread packages, Locking methods, Porting @node RTS-threads, Thread packages, Locking methods, Porting
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment