<!DOCTYPE html>
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p><font face="serif">Guys -- I packed all the cars/trucks last
night and am moving my daughter to Augusta today. I apologize
for the delay in getting the BabyBLAS back to you but it has
been difficult to find a cluster node to test your code on and
since some of you included compiler flags for features ONLY
available on Godsey 218 systems -- I generally had to wait. I
should have those finished no later than Sunday afternoon as I
have now isolated a lab218 system exclusively for my testing.</font></p>
<p><font face="serif">You may have seen that I did finish grading
the OpenMPI/MPI Proof of Concept paper. Most of you did a
fantastic job demonstrating how you could get more performance
across multiple nodes by using more threads per node. I tried
to provide some guidance for those of you that struggled with
this.</font></p>
<p><font face="serif">For the "final" you will most likely want to
create 3D plots to locate the "sweet spot". I created a video
on how to do this on CANVAS and have attached an example to this
e-mail. I will be pretty much "off the grid" until tomorrow,
but if you spot any issues with the cluster or need something go
ahead and e-mail me. If I am able to respond from my phone I
will.</font></p>
<p><font face="serif"><br>
</font></p>
<div class="moz-signature">-- <br>
<b><i>Andrew J. Pounds, Ph.D.</i></b><br>
<i>Professor of Chemistry and Computer Science</i><br>
<i>Director of the Computational Science Program</i><br>
<i>Mercer University, Macon, GA 31207 (478) 301-5627</i></div>
</body>
</html>