Abstract
We describe our experiences at parallelising a large Fortran program on a distributed-memory parallel computer. The program is the first-principles linear muffin tin orbital (LMTO) band-structure program which is used widely to calculate the electronic properties of crystalline solids, multilayers, polymers and complex molecules. The parallelising was done with (1) 3L parallel Fortran running on a quadputer, (2) Meiko's Occam Programming System (OPS) and (3) CS Tools on the Meiko Surface. We found that farming was not practicable because of livelock, the most efficient system was OPS, and that the speed of the program exceeds that of the CRAY when it is run on the larger domains of the Edinburgh Supercomputer. Also large numbers of k points can be easily handled on these domains. We give and discuss the efficiency of speed-up from our parallelisation.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
