[Wien] Parallelize Spin Up and Down (again)

[L. D. Marks]

I briefly looked into this today, and I don't think it's quite as simple 
as Peter (Sept 19th) suggested. One could get a turnaround improvement of 
2 for cases where one had, say, 4 k-points and 8 nodes available. Perhaps 
an added option -updn exploiting .machinesup & .machinesdn (rather than 
just .machines) to dispact the lapw1 (perhaps lapw2) off to the different 
nodes. Somewhat more of a rewrite of lapw1(2)para.

Note: if you have an old email address for me, please note that "nwu" has
been changed to "northwestern".
-----------------------------------------------
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60201, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
http://www.numis.northwestern.edu
-----------------------------------------------