<div>For those of us who are behind on filesystem terminology, I found "OST" = "Objective Storage Target" at <a href="http://www.ncsa.uiuc.edu/UserInfo/Resources/Hardware/News/tungsten/ByDate/Lustre_Striping.html">
http://www.ncsa.uiuc.edu/UserInfo/Resources/Hardware/News/tungsten/ByDate/Lustre_Striping.html</a></div>
<div>Peter ("a TLA without a Wiki entry is Obscure")<br><br> </div>
<div><span class="gmail_quote">On 9/12/07, <b class="gmail_sendername">Mark Hahn</b> <<a href="mailto:hahn@mcmaster.ca">hahn@mcmaster.ca</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">> I'm not sure if everyone has seen this yet, but Sun just bought Cluster<br>> File System (Lustre):
<br><br>wow, that's a bit of a shock. I had heard mutterings about CFS using ZFS,<br>but would not have guessed a buyout. where does that leave HP, for<br>instance? or even SGI, who seem to now be promoting Lustre for their
<br>new blade clusters...<br><br>the combination of ZFS and Luster is a bit strange to contemplate:<br><br> - Lustre's main thing is to set up client-OST parallel streams<br> and then get out of the way. it uses a single metadata server,
<br> which can be a bottleneck for non-giant-stream workloads, and<br> which currently stores per-file ost-mapping info via EAs in an<br> ext3 FS. OST'S don't need much functionality in their local FS,
<br> and use ext3 (with extents) as well.<br><br> - ZFS's main thing is managing storage pools directly so as to<br> provide intelligent file-level redundancy. and it has comfortingly<br> high size limits and some nice integrity features. but afaik,
<br> no provisions at all for clustering or parallel access.<br><br>I guess you could broadly compare CFS's OST's to ZFS's storage pools<br>(sorry, don't know the ZFS terminology). but "clusterizing" the metadata
<br>logic of ZFS (parallelizing it and presumably distributing it across clients)<br>sounds kind of tricky.<br><br>(I confess my only experience is with HP's SFS version of Luster, and none<br>at all with ZFS.)<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 <a href="http://www.beowulf.org/mailman/listinfo/beowulf">http://www.beowulf.org/mailman/listinfo/beowulf
</a><br></blockquote></div><br>
!DSPAM:46e83e42201372020149523!