benchmarks says something completely different...

git-svn-id: http://svn.roundup-tracker.org/svnroot/roundup/trunk@1667 57a73879-2fb5-44c3-a270-3262357dd7e2

diff --git a/doc/installation.txt b/doc/installation.txt
index 8498f319ef35c942c45a2611fa410a76f382e4f2..3dfaa77f32eebea5de0717d9bc4f60167bb3bed9 100644 (file)
--- a/doc/installation.txt
+++ b/doc/installation.txt
@@ -2,7 +2,7 @@
 Installing Roundup
 ==================
 
-:Version: $Revision: 1.47 $
+:Version: $Revision: 1.48 $
 
 .. contents::
 
@@ -240,12 +240,10 @@ There's several to choose from, each with benefits and limitations:
   There are no limitations, and it's much faster and more scaleable than the
   dbm backends.
 **mysql**
-  Backend for popular RDBMS MySQL. This backend uses MySQL's Berkeley DB
-  (transaction safe) table format, so with small amount of data it may be a
-  bit slower than \*dbm backends (due to mysql layer's overhead). However with
-  lots of data mysql may perform better (due to query optimisation). No actual
-  serious benchmarks were made though. For more info on installation of this
-  backend see doc/mysql.txt. 
+  Backend for popular RDBMS MySQL. According to benchmarks, this backend works
+  much faster than any of *dbm ones, but slightly slower than metakit and
+  sqlite. A good scalability is not a property of this backend for now,
+  though. For more info on backend installation see doc/mysql.txt. 
 **metakit**
   This backend is implemented over the metakit_ storage system, using Mk4Py as
   the interface. It scales much better than the dbm backends.