<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body dir="auto">
Firstly, we don't test anything larger than 1MB.
<div><br>
</div>
<div>Secondly, the best recordsize is up to the apication IO pattern. If it is streaming writes, it might be OK. If it is random write then probably not. <br>
<br>
<div id="AppleMailSignature">Cheers, Andreas</div>
<div><br>
On Sep 27, 2018, at 19:04, Riccardo Veraldi <<a href="mailto:Riccardo.Veraldi@cnaf.infn.it">Riccardo.Veraldi@cnaf.infn.it</a>> wrote:<br>
<br>
</div>
<blockquote type="cite">
<div>
<p>Hello,</p>
<p>I have always been using ZFS record size to 1MB with Lustre on top of ZFS. Anyway it is possible to set it to more than 1MB. May this have any benefit in performance and is it recommended to set for example:</p>
<div class="code">
<pre>echo 4194304 > /sys/module/zfs/parameters/zfs_max_recordsize</pre>
</div>
<pre>zfs set recordsize=4m zpool
</pre>
<p>Thank you.</p>
<p>Rick</p>
<p><br>
</p>
</div>
</blockquote>
<blockquote type="cite">
<div><span>_______________________________________________</span><br>
<span>lustre-discuss mailing list</span><br>
<span><a href="mailto:lustre-discuss@lists.lustre.org">lustre-discuss@lists.lustre.org</a></span><br>
<span><a href="http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org">http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org</a></span><br>
</div>
</blockquote>
</div>
</body>
</html>