pgpool2.git / summary / log / commit / refs

commit    a7aa896643fcd6ea8dbcd718b87922a91d3fb749
Author:   Tatsuo Ishii <ishii@sraoss.co.jp>
Date:     Fri Oct 07 02:26:28 2022 +0000

    Doc: enhance description about memqcache_method.
    
    Add explanation which method should be used.
    
    Backpatch-through: 3.7.


 doc.ja/src/sgml/memcache.sgml | 27 +++++++++++++++++++++++++++
 doc/src/sgml/memcache.sgml    | 18 +++++++++++++++++-
 2 files changed, 44 insertions(+), 1 deletion(-)

diff --git a/doc.ja/src/sgml/memcache.sgml b/doc.ja/src/sgml/memcache.sgml
index bd5284891..07bf2c99d 100644
--- a/doc.ja/src/sgml/memcache.sgml
+++ b/doc.ja/src/sgml/memcache.sgml
@@ -218,6 +218,33 @@
       </tgroup>
      </table>
 
+     <para>
+<!--
+      In general <varname>shmem</varname> is faster
+      than <varname>memcached</varname> because the former does not
+      involve any network access. However, the upper limit size of
+      <varname>memqcache_total_size</varname> is limited by the total
+      shared memory size defined by the system. The size is large
+      enough in modern Linux systems, but in other systems there maybe
+      smaller upper limit. In this case you need to change the system
+      settings regarding the maximum shared memory size. There's a
+      good explanation about shared memory
+      in <ulink url="https://www.postgresql.org/docs/current/kernel-resources.html#SYSVIPC"><productname>PostgreSQL</productname>
+      documentation</ulink>.
+-->
+      一般的に<varname>shmem</varname>はネットワークアクセスが不要なため、<varname>memcached</varname>より高速です。
+      しかし、<varname>shmem</varname>では、<varname>memqcache_total_size</varname>はシステムで決まっている共有メモリサイズの合計に制限されます。
+      最近のLinuxシステムでは、そのサイズは十分大きいのですが、他のシステムでは制限値が小さいかもしれません。
+      この場合は共有メモリサイズの最大値のシステム設定を変更する必要があるでしょう。
+      <ulink url="https://www.postgresql.org/docs/current/kernel-resources.html#SYSVIPC"><productname>PostgreSQL</productname></ulink>のドキュメントに共有メモリに関する良い記述があります。
+     </para>
+     <para>
+<!--
+      If you are not sure which memqcache_method to be used, start with <varname>shmem</varname>.
+-->
+      どちらのmemqcache_methodを使ったらよいかわからない場合は、まず<varname>shmem</varname>を試してください。
+     </para>
+
      <para>
       <!--
       Default is <literal>'shmem'</literal>.
diff --git a/doc/src/sgml/memcache.sgml b/doc/src/sgml/memcache.sgml
index 2e350c59a..995a52fda 100644
--- a/doc/src/sgml/memcache.sgml
+++ b/doc/src/sgml/memcache.sgml
@@ -133,7 +133,7 @@
      </para>
 
      <table id="memqcache-method-table">
-      <title>Memcache method options</title>
+      <title>Memqcache method options</title>
       <tgroup cols="2">
        <thead>
 	<row>
@@ -157,6 +157,22 @@
       </tgroup>
      </table>
 
+     <para>
+      In general <varname>shmem</varname> is faster
+      than <varname>memcached</varname> because the former does not
+      involve any network access. However, the upper limit size of
+      <varname>memqcache_total_size</varname> is limited by the total
+      shared memory size defined by the system. The size is large
+      enough in modern Linux systems, but in other systems there maybe
+      smaller upper limit. In this case you need to change the system
+      settings regarding the maximum shared memory size. There's a
+      good explanation about shared memory
+      in <ulink url="https://www.postgresql.org/docs/current/kernel-resources.html#SYSVIPC"><productname>PostgreSQL</productname>
+      documentation</ulink>.
+     </para>
+     <para>
+      If you are not sure which memqcache_method to be used, start with <varname>shmem</varname>.
+     </para>
      <para>
       Default is <literal>'shmem'</literal>.
      </para>


[parent: 13f22bcfa8c4]