TURBOMOLE Users Forum
Installation and usage of TURBOMOLE => Parallel Runs => Topic started by: f.uhlig on December 17, 2010, 05:38:43 PM
-
Dear all,
I was wondering whether there is a plan to make the "$incore <memsize>" option available for parallel execution of the self-consistent field methods?
I have so far seen a significant influence on the performance of especially the dscf-module in sequential execution. A parallel implementation would be very helpful, not only on high-performance clusters, but also on smaller Linux computer clusters that get used very widely....
Best regards,
Frank Uhlig
-
maybe, if there is a volunteer who has the time and the knowhow to do the work...
Christof