Project: Memcached

Revision: 794

Author: dormando

Date: 07 Sep 2008 20:21:11

• /trunk/website/index.bml

  ......	@@ -35,7 +35,7 @@
  35	35	<p>If a host goes down, the API re-maps that dead host's requests onto the servers that are available.</p>
  36	36
  37	37	<?h1 Shouldn't the database do this? h1?>
  38		<p>Regardless of what database you use (MS-SQL, Oracle, Postgres, MysQL-InnoDB, etc..), there's a lot of overhead in implementing <a href="http://www.wikipedia.org/wiki/ACID">ACID</a> properties in a RDBMS, especially when disks are involved, which means queries are going to block. For databases that aren't ACID-compliant (like MySQL-MyISAM), that overhead doesn't exist, but reading threads block on the writing threads.</p>
  	38	<p>Regardless of what database you use (MS-SQL, Oracle, Postgres, MySQL-InnoDB, etc..), there's a lot of overhead in implementing <a href="http://www.wikipedia.org/wiki/ACID">ACID</a> properties in a RDBMS, especially when disks are involved, which means queries are going to block. For databases that aren't ACID-compliant (like MySQL-MyISAM), that overhead doesn't exist, but reading threads block on the writing threads.</p>
  39	39	<p><?memd?> never blocks. See the "Is memcached fast?" question below.</p>
  40	40
  41	41	<?h1 What about shared memory? h1?>
  ......	@@ -76,7 +76,7 @@
  76	76	but you can't spread writes (they have to process on all machines) and
  77	77	they'll eventually consume all your resources. You'll find yourself
  78	78	adding replicated slaves at an ever-increasing rate to make up for the
  79		diminishing returns each addition slave provides.</p>
  	79	diminishing returns each additional slave provides.</p>
  80	80
  81	81	<p>The next logical step is to horizontally partition your dataset
  82	82	onto different master/slave clusters so you can spread your writes,
  ......	@@ -118,4 +118,4 @@
  118	118	</ul>
  119	119
  120	120	<=body
  121		page?>
  122	121	\ No newline at end of file
  	122	page?>