<html><body>
<p><tt><font size="2">Hi Orlando,</font></tt><br>
<br>
<tt><font size="2">Thanks for all of your feedback, many great suggestions. Sorry for the late response, I've been trying to go through and digest all the comments from the user group meeting. I'll do my best to forward your suggestions internally.</font></tt><br>
<br>
<tt><font size="2">The one thing I wanted to comment on was that "hot file" identification was shipped in gpfs 3.5.0.3.</font></tt><br>
<br>
<tt><font size="2">Here is a link to the docs discussing it:</font></tt><br>
<tt><font size="2"><a href="http://publib.boulder.ibm.com/infocenter/clresctr/vxrx/index.jsp?topic=%2Fcom.ibm.cluster.gpfs.v3r50-3.gpfs200.doc%2Fbl1adv_userpool.htm&resultof=%22file%22%20%22heat%22%20">http://publib.boulder.ibm.com/infocenter/clresctr/vxrx/index.jsp?topic=%2Fcom.ibm.cluster.gpfs.v3r50-3.gpfs200.doc%2Fbl1adv_userpool.htm&resultof=%22file%22%20%22heat%22%20</a></font></tt><br>
<br>
<br>
<tt><font size="2">Dean Hildebrand<br>
Research Staff Member - Storage Systems<br>
IBM Almaden Research Center</font></tt><br>
<br>
<tt><font size="2">On 25/09/12 14:05, Jez Tucker wrote:<br>
></font></tt><tt><font size="2"><i> Hello all<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> Firstly can I thank all who attended UG #6. We had a great turn out and the opportunity to network with more people from IBM was most welcome.<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> I have uploaded the presentations from UG to this small, catchy URL: </i></font></tt><a href="http://goo.gl/n1in1"><tt><font size="2" color="#0000FF"><i><u>http://goo.gl/n1in1</u></i></font></tt></a><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> [Bar the SCCS presentation, awaiting clearance].<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> Please have a read of the presentations.<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> IBM Almaden Labs welcome your feedback regarding pNFS and Panache as well as FRQs etc.<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> For instance, one FRQ idea banded around was a GRIO/QoS implementation for GPFS : E.G.: </i></font></tt><a href="http://goo.gl/zjkN8"><tt><font size="2" color="#0000FF"><i><u>http://goo.gl/zjkN8</u></i></font></tt></a><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> It would be most helpful if a couple of lines use-case was alongside each of these.<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> If these are messaged back to the list for healthy debate or sent to me directly I'll put them on the UG website for Almaden Labs to peruse/discuss with us.<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> Regards,<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> Jez<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i><br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> p.s. I'll also start to solicit previous presentations for UG < #5, so if you were a speaker, please get in touch.<br>
</i></font></tt><tt><font size="2">></font></tt><tt><font size="2"><i> ---<br>
</i></font></tt><tt><font size="2"><br>
Thanks for the great meeting Jez, and Claire et al at OCF.<br>
<br>
On feature requests, I think one desirable feature request discussed at <br>
the meeting was for "better" performance monitoring tools.<br>
<br>
A quick think through the things on my plate which would be eased with <br>
new/changed features in GPFS led me to this wishlist:<br>
<br>
- ability to change the designated NSD servers for an NSD without <br>
unmounting the filesystem everywhere<br>
<br>
- expansion of the AFM toolchain, including the following to assist <br>
with migration of data between filesystems:<br>
- ability to set a pre-existing fileset as a "cache" of an empty <br>
'home' fileset with AFM, allowing for a push of the data from the <br>
"cache" fileset/filesystem to the "home" target fileset/filesystem as a <br>
data migration strategy<br>
- ability to remove an AFM relationship between filesets, preserving <br>
data in the 'cache' fileset (and making it, independently, a 'live' fileset)<br>
- ability to "flip" the 'home'<->'cache' relationship, resulting in <br>
a flush from the new 'cache' fileset to the new 'home' fileset<br>
<br>
- better documentation (and, indeed, automation/automagic) on making <br>
best use of available memory within NSD servers<br>
<br>
- read caching of data blocks within an NSD server's memory (when <br>
acting in "server" mode in a multi-cluster environment where the client <br>
nodes do not have direct block access to the disks)<br>
<br>
- "hot file" identification tools/data for policy based HSM migration<br>
<br>
- some easy and non-invasive method for logging file and folder <br>
deletions (for the purposes of expiring backup data without using a <br>
separate database of files, in my case)<br>
<br>
- better licensing model (dare I say it - capacity based?)<br>
<br>
<br>
I'd love to be able to change the blocksize on an existing filesystem <br>
too, but I imagine that's not possible.<br>
<br>
<br>
--<br>
Orlando<br>
<br>
<br>
<br>
-- <br>
--<br>
Dr Orlando Richards<br>
Information Services<br>
IT Infrastructure Division<br>
Unix Section<br>
Tel: 0131 650 4994<br>
<br>
The University of Edinburgh is a charitable body, registered in <br>
Scotland, with registration number SC005336.<br>
</font></tt><br>
<br>
<tt><font size="2"><br>
</font></tt></body></html>