[Wien] Problem with testpara
Wien2k User
wien2k.user at gmail.com
Tue Feb 21 19:00:59 CET 2017
Thank you very much for your answer
I used this configuration because the nodes do not have the same
performance (one machine faster than another)
I copied lapwpara and it works very well but I would be grateful if you
could explain to me how to use 'balance' with the new file lapwpara
2017-02-21 15:01 GMT+01:00 Wien2k User <wien2k.user at gmail.com>:
> Thank you very much for your answer
>
> I tried this change but it gives the same thing (no change is made) but
> with wien2k_14 it changes and it does not give the same thing relatively to
> version 16.
>
> 2017-02-21 14:39 GMT+01:00 Wien2k User <wien2k.user at gmail.com>:
>
>> Dear Wien2k users,
>>
>> I am still waiting for an answer from Prof P.Blaha or Wien2k users.
>>
>> 2017-02-20 1:28 GMT+01:00 Wien2k User <wien2k.user at gmail.com>:
>>
>>> Dear Prof P.Blaha,
>>>
>>> I launched a calculation with 126 k-point under a cluster but when I run
>>> the testpara_lapw with wien2k_16 and wien2k_14 with the same machine file
>>> below, the first version attribute 6 k-point for each thread of machine-1
>>> and 5 k-point for machine-2 and machine-3 but version 14 assigns 8 k-point
>>> for machine-1 and 3 k-point for machine-2 and machine-3. The two versions
>>> do not give the same thing and I prefer the distribution of version 14. how
>>> to apply this distribution with version 16?
>>>
>>>
>>> .machines:
>>>
>>> 2 :machine-1
>>>
>>> 2 :machine-1
>>>
>>> 2 :machine-1
>>>
>>> 2 :machine-1
>>>
>>> 2 :machine-1
>>>
>>> 2 :machine-1
>>>
>>> 2 :machine-1
>>>
>>> 2 :machine-1
>>>
>>> 2 :machine-1
>>>
>>> 2 :machine-1
>>>
>>> 1 :machine-2
>>>
>>> 1 :machine-2
>>>
>>> 1 :machine-2
>>>
>>> 1 :machine-2
>>>
>>> 1 :machine-3
>>>
>>> 1 :machine-3
>>>
>>> 1 :machine-3
>>>
>>> 1 :machine-3
>>>
>>> 1 :machine-3
>>>
>>> 1 :machine-3
>>>
>>> 1 :machine-3
>>>
>>> 1 :machine-3
>>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20170221/e802cfe3/attachment.html>
More information about the Wien
mailing list