<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">That's sort of what I did. <div>I ran the program for about 20 minutes and checked my Mac's Activity Monitor under Utilities. Ian Taylor was right that <span class="Apple-style-span" style="font-family: arial, sans-serif; ">745 GB was way too much to ask for.</span><span class="Apple-style-span" style="font-family: arial, sans-serif; font-size: 12px; "> </span>The program used up to 2.3GB at a time so I asked UF's cluster for a 4GB node. I know this sounds like a ridiculous amount, but the model is recursive with up to 7374 sums needed to predict each data point with nearly 7000 data points.</div><div><div><div apple-content-edited="true"><br class="Apple-interchange-newline"> </div><div apple-content-edited="true">Thanks for all the advice!</div><br><div><div>On Sep 11, 2010, at 2:13 PM, dave fournier wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div>Actually it would not be too difficult to have ADMB do all this automatically.<br>One could do one "fake" function evaluation and keep track of the amount of memory<br>buffers needed and allocate them. The user could optionally input a maximum amount of<br>memory to be used. that was what I had intended to do ... one day.<br>_______________________________________________<br>Users mailing list<br><a href="mailto:Users@admb-project.org">Users@admb-project.org</a><br>http://lists.admb-project.org/mailman/listinfo/users<br><br></div></blockquote></div><br></div></div></body></html>