<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Dear PETSc Experts!<br>
<br>
Do you know if there is an efficient way to move a matrix from a
single processor (MatCreateSeqBAIJ) to a matrix contained in all
processors?<br>
<br>
As a little bit of context, I have a code in which only one
processor creates a matrix and a vector for a linear system of
equations. Then we want to use a parallel solver to get the solution
and give it back to a single processor<br>
<br>
I tried MatView to create a binary file and MatLoad to load the
matrix in parallel. This seems to work but performance is
significantly decreased independent of the number of processors
used.<br>
<br>
I have some questions:<br>
<br>
Can I share the matrix without having to write it to a file, for
example, through a buffer?<br>
Is there a way to efficiently avoid the overhead of writing, reading
loading matrices to and from processors?<br>
<br>
Thanks for your comments,<br>
<br>
Hector<br>
<div class="moz-signature">-- <br>
<font size="2"><b>Hector Barrios</b><br>
PhD Student, Graduate Research Assistant<br>
Hildebrand Department of Petroleum and Geosystems Engineering<br>
<font color="#bf5700">The University of Texas at Austin</font><br>
<a href="mailto:hectorb@utexas.edu" target="_blank">hectorb@utexas.edu</a></font><br>
</div>
</body>
</html>