<br><div class="gmail_quote">On Sun, Jul 17, 2011 at 4:14 PM, Mihael Hategan <span dir="ltr"><<a href="mailto:hategan@mcs.anl.gov">hategan@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;">
<div class="im">On Sun, 2011-07-17 at 16:10 -0500, Ketan Maheshwari wrote:<br>
><br>
><br>
> On Sun, Jul 17, 2011 at 2:07 PM, Mihael Hategan <<a href="mailto:hategan@mcs.anl.gov">hategan@mcs.anl.gov</a>><br>
> wrote:<br>
> On Sun, 2011-07-17 at 11:59 -0700, Mihael Hategan wrote:<br>
> > ><br>
> > > Theres a special issue of J. Grid Computing call for<br>
> data-intensive<br>
> > > Cloud papers due Aug 16 that might be good for this<br>
> (perhaps stressing<br>
> > > the provide-staging aspects)?<br>
> ><br>
> > Sounds good. I'll start getting some of the necessary<br>
> numbers.<br>
><br>
><br>
> Which reminds me: who wants to help with said numbers?<br>
><br>
> For example, there is an easy one: the dependence between job<br>
> wall time<br>
> and queuing time on various clusters.<br>
><br>
> I can try this along with the cybershake work I am doing these days.<br>
<br>
</div>I'm not sure where cybershake fits in. The problem would be to submit<br>
dummy jobs of various walltimes to various clusters and record the<br>
amount of time they sit in the queue.<br></blockquote><div> </div><div>It is just a use-case. I thought, stats based on an (any) real application will have more merit compared to dummy jobs.<br><br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<br>
Though an alternative may be to go through swift logs and extract that<br>
data. It may not give us relevant data points, but it may be worth a<br>
try.<br>
<div class="im">><br>
> I read the text of the paper (in March) and have some comments, will<br>
> send'em soon.<br>
><br>
</div>Thanks.<br>
<br>
<br>
</blockquote></div><br><br clear="all"><br>-- <br>Ketan<br><br><br>