<div>In my humble opinion, I think you can start with NFS (but using at least NFSv4), and see what happens, for instance, what's the real disk access pattern of your cluster, in terms of amount of IOPS, average BW usage, and read/write pattern.
<div>It could be interesting to know not only what's gonna be the file server, but what's the underlying storage subsystem, and the network requirements. I mean, the server could be big, the network sth like IB or Myrinet, and then you can be serving files that reside on a copule of internal disks so performance would sink...
<div>Anyway, I've seen some NFS file servers in that cluster size, I'd suggest try it and check if this is enough for your applications. </div>
<div>Hope this helps,</div>
<div><span class="gmail_quote">2007/3/1, <a href="mailto:firstname.lastname@example.org">email@example.com</a> <<a href="mailto:firstname.lastname@example.org">email@example.com</a>>:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">Hi,<br><br>I have a small (16 dual xeon machines) cluster. We are going to add<br>an additional machine which is only going to serve a big filesystem via
<br>a gigabit interface.<br><br>Does anybody knows what is better for a cluster of this size, exporting the<br>filesystem via NFS or use another alternative such as a cluster filesystem<br>like GFS or OCFS?<br><br>Thanks in advance
<br><br>--<br><br> Jaime D. Perea Duarte. <jaime at iaa dot es><br> Linux registered user #10472<br><br> Dep. Astrofisica Extragalactica.<br> Instituto de Astrofisica de Andalucia (CSIC)
<br> Apdo. 3004, 18080 Granada, Spain.<br>_______________________________________________<br>Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org">Beowulf@beowulf.org</a><br>To change your subscription (digest mode or unsubscribe) visit