Mike,<div><br></div><div>Here is an old mail snippet (May) that you sent about the scratch tag:</div><div><br></div><div>"</div><div><span class="Apple-style-span" style="font-family: arial, sans-serif; font-size: 13px; background-color: rgb(255, 255, 255); ">I was surprised to see that coaster provider staging used the <workdirectory> tag to determine the jobdir on the compute node, on Beagle where /tmp is not writeable. I always thought that it would honor the <<span class="il" style="background-image: initial; background-attachment: initial; background-origin: initial; background-clip: initial; background-color: rgb(255, 255, 136); color: rgb(34, 34, 34); background-position: initial initial; background-repeat: initial initial; ">scratch</span>> tag to let the user specify the provider staging jobdir. But this seems not to be the case. </span><br>
"</div><div><br></div><div>So, it seems that the scratch tag is not really honored.</div><div><br></div><div><br></div><div><div class="gmail_quote">On Tue, Aug 16, 2011 at 10:08 PM, Michael Wilde <span dir="ltr"><<a href="mailto:wilde@mcs.anl.gov">wilde@mcs.anl.gov</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">Right, I remember that now, but I cant recall if the scratch tag was completely broken, was set incorrectly, was interfering with debugging, or was interacting incorrectly with some other setting.<br>
<br>
Papia, if you recall, please clarify.<br>
<br>
Thanks,<br>
<br>
- Mike<br>
<div class="im"><br>
----- Original Message -----<br>
> From: "Jonathan Monette" <<a href="mailto:jonmon@mcs.anl.gov">jonmon@mcs.anl.gov</a>><br>
> To: "Michael Wilde" <<a href="mailto:wilde@mcs.anl.gov">wilde@mcs.anl.gov</a>><br>
</div><div><div></div><div class="h5">> Cc: "swift-devel Devel" <<a href="mailto:swift-devel@ci.uchicago.edu">swift-devel@ci.uchicago.edu</a>><br>
> Sent: Tuesday, August 16, 2011 9:14:24 PM<br>
> Subject: Re: provider staging question<br>
> That is what I was thinking about CDM but wasn't sure if that is what<br>
> you meant in the email a couple days ago.<br>
><br>
> I thought that the scratch tag was broken from the sites file. I<br>
> thought it was mentioned before that Papia was experiencing problems<br>
> when using the scratch tag in her sites file.<br>
><br>
> On Aug 16, 2011, at 8:15 PM, Michael Wilde wrote:<br>
><br>
> > Hi Jon,<br>
> ><br>
> > I'll try a short answer, but this needs more thoight, and much<br>
> > testing:<br>
> ><br>
> > - Dont use CDM with provider staging (yet; maybe someday that will<br>
> > make sense...)<br>
> > I dont tthink the two will work together well.<br>
> ><br>
> > - ps should stage to the local hard disk; I *think* it may honor the<br>
> > workdirectory tag as where to stage, though. Or maybe the scratch<br>
> > tag? I think it defaults to /tmp. This needs to be tested and<br>
> > documented. Mihael, can you clarify?<br>
> ><br>
> ><br>
> > Miek<br>
> ><br>
> ><br>
> > ----- Original Message -----<br>
> >> From: "Jonathan Monette" <<a href="mailto:jonmon@mcs.anl.gov">jonmon@mcs.anl.gov</a>><br>
> >> To: "Michael Wilde" <<a href="mailto:wilde@mcs.anl.gov">wilde@mcs.anl.gov</a>><br>
> >> Sent: Tuesday, August 16, 2011 6:08:47 PM<br>
> >> Subject: provider staging question<br>
> >> Mike,<br>
> >> I am configuring the run to test the SwiftMontage runs with<br>
> >> provider<br>
> >> staging turned on as you suggested. When I turn on provider staging<br>
> >> should I not use CDM? The issue that I was experiencing that led me<br>
> >> to<br>
> >> start using CDM was extremely long copy times from the cwd to the<br>
> >> job<br>
> >> directory specified by the <workdirectory> in the sites.xml file.<br>
> >> Does<br>
> >> provider staging circumvent that issue? I know provider staging<br>
> >> copies<br>
> >> the input files directly onto the compute nodes local disk but that<br>
> >> is<br>
> >> about all I know. Could you fill me in a bit on what exactly<br>
> >> provider<br>
> >> staging does? Would using both CDM "direct" directives and provider<br>
> >> staging cause some degraded performance?<br>
> ><br>
> > --<br>
> > Michael Wilde<br>
> > Computation Institute, University of Chicago<br>
> > Mathematics and Computer Science Division<br>
> > Argonne National Laboratory<br>
> ><br>
<br>
--<br>
Michael Wilde<br>
Computation Institute, University of Chicago<br>
Mathematics and Computer Science Division<br>
Argonne National Laboratory<br>
<br>
_______________________________________________<br>
Swift-devel mailing list<br>
<a href="mailto:Swift-devel@ci.uchicago.edu">Swift-devel@ci.uchicago.edu</a><br>
<a href="https://lists.ci.uchicago.edu/cgi-bin/mailman/listinfo/swift-devel" target="_blank">https://lists.ci.uchicago.edu/cgi-bin/mailman/listinfo/swift-devel</a><br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br>Ketan<br><br><br>
</div>