Makefile for Xeon 2 linux cluster mkl/10.0.011/lib/em64t

[ravi]

Hello,

I am trying to install vasp4.6.31 in a linux cluster contain Xeon 2
i.e. http://www.notur.no/hardware/stallo/ ). Though I am able to create vasp executable the job fail for test case (e.g. fcc Cu) . I am getting strange total energy values and hence after
4 scf cycle getting an errormessage

## N E dE d eps ncg rms rms(c)
DAV: 1 0.282009941511E+02 0.28201E+02 -0.64019E+03 1698 0.177E+03
DAV: 2 -0.947871324610E+01 -0.37680E+02 -0.35331E+02 1688 0.194E+02
DAV: 3 -0.101411396801E+02 -0.66243E+00 -0.66004E+00 1896 0.311E+01
DAV: 4 -0.101425808650E+02 -0.14412E-02 -0.14411E-02 1820 0.136E+00
DAV: 5 -0.101425829177E+02 -0.20527E-05 -0.20523E-05 1794 0.495E-02BRMIX: very serious
problems
the old and the new charge density differ
old charge density: 11.36486 new 11.30349
0.921E+00
#
For the same job running without any problem in another linux cluster
(http://www.notur.no/hardware/snowstorm/) where the result from first 4 scf cycle for the same job is
N E dE d eps ncg rms rms(c)
DAV: 1 0.339325358500E+02 0.33933E+02 -0.64365E+03 1698 0.178E+03
DAV: 2 -0.394495003525E+01 -0.37877E+02 -0.35438E+02 1688 0.195E+02
DAV: 3 -0.457899938445E+01 -0.63405E+00 -0.63248E+00 1878 0.301E+01
DAV: 4 -0.458043509660E+01 -0.14357E-02 -0.14356E-02 1818 0.135E+00
#

You can notice that there is signeficant difference between the tow output though they are from same input. I have spent last one week to resolve this problem and could not find a solution to obtain working version of vasp.

Could you kindly give me a makefile that will work with Xeon2, mkl 10.0.01/em64t ifort compiler?

Best regards
Ravi

[admin]

please make sure that hyper-threading is explicitely switched off, this may be one of the main sources of errors on Xeon2 clusters