<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
<tt>Thank you Andreas and Ashley.<br>
<br>
Together your answers really helped.<br>
Though the file creation rates did only improve by 2x, all file
listing operations are now about 20x as fast.<br>
The problem was that ACLs were not activated on the MDS but
requested by default aliased "ls". Every single line of "ls -l"
returned a "Operation not suported" error message with strace and
took about 100-300ms.<br>
<br>
Thanks again!<br>
<br>
Robert<br>
<br>
<br>
<br>
</tt><br>
Am 12.08.2010 18:22, schrieb Andreas Dilger:
<blockquote
cite="mid:156E6ADC-8B49-4787-B83E-DA563DD1FF10@oracle.com"
type="cite">
<pre wrap="">On 2010-08-12, at 10:10, robert wrote:
</pre>
<blockquote type="cite">
<pre wrap="">I am trying to solve a problem with very slow "ls" and other big amount
of file operations but good overall read/write rates.
We are running a small cluster of 3 OSSs with 9 OSTs, 1MDS (with SSD
MDT) and currently two clients. All server nodes are centos 5.2 with
lustre 1.8.1 while the clients are centos 5.4 with lustre 1.8.3. All
components are networked with DDR IB. Striping is set to 1 or 2 for
different folders.
</pre>
</blockquote>
<pre wrap="">
I don't recall if there are specific metadata improvements in 1.8.3 vs. 1.8.1, but it is usually better to have newer RPMs than older. Note that if you have 1.8.1 (vs. 1.8.1.1) there is a corruption bug in the MDS, and you definitely need to upgrade.
</pre>
<blockquote type="cite">
<pre wrap="">From the very beginning of our tests we had rather slow metadata
operations. File creation maxed at 250/s/client.
</pre>
</blockquote>
<pre wrap="">
For DDR IB that is definitely slow.
</pre>
<blockquote type="cite">
<pre wrap="">"ls" of a dir with 1000
files takes about 40-70 seconds almost independently from the file´s
sizes. Dirs that have recently been accessed, are of course much faster
due to caching. There is no general performance problem as we are
getting almost 1G/s when reading/writing big files from two clients in
several threads. But when creating lots of files with lmdd in a test
script in a single thread there are also hangs of a few seconds before
the rates get back to normal.
I´ve been searching the mailing list archives for similar problems but
only found the usual "improving performance for small files" hints. All
these suggestions have been tested but did not or only slightly improve
performance. Can anyone please tell me ...
- Is there a way to check the amount of time that the different parts of
a file operation take (like 1ms requesting metadata, 1ms receiving
metadata, 123ms reading blocks from OST, ...)?
</pre>
</blockquote>
<pre wrap="">
You can use "strace -ttt" on the client to see the time of each syscall. You can look at the various "stats" files on the MDS to see the wait+processing time of each RPC type.
Cheers, Andreas
--
Andreas Dilger
Lustre Technical Lead
Oracle Corporation Canada Inc.
</pre>
</blockquote>
<br>
<div class="moz-signature">-- <br>
<title></title>
<p style="line-height: 1.2em;"><font face="Microsoft Sans Serif"><small><font
color="#333333"><b><small>Robert
Pinnow - Managing Director - <font face="Verdana">r i
s e | fx</font></small></b></font></small>
</font></p>
<p style="line-height: 1.1em;">
<font face="Microsoft Sans Serif"><b><font color="#666666"><small><small>t:
+49
30 201 803 00 </small></small></font></b></font><b><font
face="Microsoft Sans Serif"><font color="#666666"><small><small>robert
@
risefx.com </small></small></font></font></b><font
face="Microsoft Sans Serif"><b><font color="#666666"><small><small><br>
c: +49 172 384 2183 <a class="moz-txt-link-abbreviated" href="http://www.risefx.com">www.risefx.com</a></small></small></font></b><br>
</font></p>
<p style="line-height: 1.1em;">
<font face="Microsoft Sans Serif"><small><b><small><font
color="#333333" face="Verdana">r i s e | fx GmbH</font><font
face="Verdana"> </font></small><br>
<small><font color="#666666">Schlesische Strasse 28,
Aufgang B 10997
Berlin</font></small></b></small><b><font
color="#666666"><br>
<small><small>Geschaeftsfuehrer: Sven Pannicke, Robert
Pinnow <br>
Handelsregister Berlin HRB 106667 B</small></small></font></b></font></p>
</div>
</body>
</html>