[Wien] machines file and klist splitting

[Stefaan Cottenier]

> I setup a .machines file, and I'm suspicious of the how the klist is
> being 
> split.  It seems non-optimal, so I suspect I did something wrong.

You didn't make a mistake. But for really optimal performance, you 
need to tune the .machines file case by case. There are two cures for 
your case:

1) using a residue machine:

## begin .machines file ##
granularity:1
1:node001:1
1:node001:1
1:node002:1
1:node002:1
residue:node003
## end .machines file ##

The price is that you have to use an extra node. Ideally, you use a 
slower node for this (if you have one), as it will take that 2 k-
points, and hence can be 5 times as slow as the other nodes, without 
costing extra time.

2) distribute the k-points yourself:
 
## begin .machines file ##
granularity:1
11:node001:1
11:node001:1
10:node002:1
10:node002:1
## end .machines file ##

Now you don't need an extra machine, but node002 will be idle for a 
short while.

Stefaan