<br><br><div class="gmail_quote">2009/7/31 Harald Stürzebecher <span dir="ltr"><<a href="mailto:haralds@cs.tu-berlin.de">haralds@cs.tu-berlin.de</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hello!<br>
<br>
2009/7/31 Yi Ling <<a href="http://lingyi.pro" target="_blank">lingyi.pro</a>@<a href="http://gmail.com" target="_blank">gmail.com</a>>:<br>
<div class="im">> there is a section about metadata on page <a href="http://www.gluster.com/news.php" target="_blank">http://www.gluster.com/news.php</a><br>
><br>
> "No metadata dependencies - metadata management is frequently a source of<br>
> performance bottlenecks, particularly when there are a very large number of<br>
> files, and can be a single point of failure in traditional file systems.<br>
> GlusterFS eliminates this problem with intelligent data placement, an<br>
> automatic data distribution method that optimizes the use of resources and<br>
> eliminates metadata bottlenecks."<br>
><br>
> yes,it's a advantage of GlusterFS. but yet i found no more detailed<br>
> information about the "intelligent data placement" "an automatic data<br>
> distribution methed". could anyone explain it in detail? and something about<br>
> file managemment~~~<br>
<br>
</div>AFAIK, the GlusterFS metadata is stored as extended attributes to the<br>
files in the underlying filesystem.<br>
<br>
Hints can be found here:<br>
<br>
<a href="http://www.gluster.org/docs/index.php/Understanding_AFR_Translator#Locking_and_Change_Log" target="_blank">http://www.gluster.org/docs/index.php/Understanding_AFR_Translator#Locking_and_Change_Log</a><br>
<br>
<a href="http://www.gluster.org/docs/index.php/Setting_up_AFR_on_two_servers_with_pre-existing_data#Preparing_the_directories" target="_blank">http://www.gluster.org/docs/index.php/Setting_up_AFR_on_two_servers_with_pre-existing_data#Preparing_the_directories</a><br>
<font color="#888888"><br>
<br>
Harald Stürzebecher<br>
</font></blockquote></div><br>thanks for replying~~ but im still confused~~~<br><br>in client, what's the routine of an operation like "create"/"read" ,from client to server (even backend filesystem)? <br>
<br>"the GlusterFS metadata is stored as extended attributes to the files in the underlying filesystem"<br>well, after storing, how could i read the file and how to know where the file is located (or which server/brick the file is stored)<br>
<br>