[CP2K-user] [CP2K:12283] Running Cp2k in parallel using thread in a PC

[Travis]

Hi,

Vibrational analysis is done numerically for each mode. With many atoms, it 
takes a very long time. NPROC_REP requests that you split this job into 
smaller segments and use N processors for each of those tasks. For example,

mpiexec_mpt -np 2304 cp2k.7.0.psmp foo.inp > foo.log

with,

   NPROC_REP  576

dumps 4 files (numbered 0 to 3). These files follow the SCF procedure for 
each calculation. The final spectral data is written to the main output 
file (foo.log, above). This is written at the very end, not piecewise. You 
should also get a Molden format file for visualizing the spectrum. I'm 
using OMP_NUM_THREADS set to 1 in the example above. For jobs on a single 
node, I'd go with OMP_NUM_THREADS equal 1 when you already parallelize your 
calculations by splitting them up like this.

-T


On Monday, September 30, 2019 at 12:38:32 AM UTC-3, Chn wrote:
>
> Hi,
> The keyword NPROC_REP has a default value of {1}, it should be only one 
> output *-r.out file in a job. So can I explain that I got six files just 
> because I used six threads..? Is it normal in a parallel job..?
> regards,
> chn
>
> 在 2019年9月29日星期日 UTC+8上午3:08:34，Ari Paavo Seitsonen写道：
>>
>> Hello,
>>
>>   Maybe you received the six files because of this:
>>
>> <<
>>
>> NPROC_REP {Integer} 
>>
>> *Specify the number of processors to be used per replica environment (for 
>> parallel runs). In case of mode selective calculations more than one 
>> replica will start a block Davidson algorithm to track more than only one 
>> frequency*  [Edit on GitHub 
>> <https://github.com/cp2k/cp2k/blob/master/src/motion/input_cp2k_vib.F#L83>] 
>>
>>
>> This keyword cannot be repeated and it expects precisely one integer. 
>>
>> Default value: 1>>
>>
>>
>> https://manual.cp2k.org/trunk/CP2K_INPUT/VIBRATIONAL_ANALYSIS.html#list_NPROC_REP
>>
>>     Greetings from Paris,
>>
>>        apsi
>>
>> Le sam. 28 sept. 2019 à 14:14, Pierre-André Cazade <pie... at gmail.com> 
>> a écrit :
>>
>>> Hi Nikhil,
>>>
>>> As you are using a mix of MPI and OpenMP, you have to use the executable 
>>> with the extension psmp.
>>>
>>> You can find a table describing all the executables in the section 3 of 
>>> the “how to compile” page:
>>>
>>> https://www.cp2k.org/howto:compile
>>>
>>> Yet, it does not explain why your calculation behaved as if there were 6 
>>> independent calculations.
>>>
>>> Please try the same calculation with the psmp executable and let me know 
>>> how it goes.
>>>
>>> Regards,
>>> Pierre 
>>>
>>> Get Outlook for iOS <https://aka.ms/o0ukef>
>>>  
>>> ------------------------------
>>> *From:* c... at googlegroups.com on behalf of Chn <ch... at gmail.com>
>>> *Sent:* Saturday, September 28, 2019 3:03 a.m.
>>> *To:* cp2k
>>> *Subject:* Re: [CP2K:12283] Running Cp2k in parallel using thread in a 
>>> PC 
>>>  
>>> Hi Pierre, 
>>> I tried to combine openMP with MPI as you mentioned above when I do 
>>> vibrational analysis. I required 6 MPI threads and got 6 output files named 
>>> as *-r-number.out, however in each file it printed that:
>>>  GLOBAL| Total number of message passing processes                      
>>>        1
>>>  GLOBAL| Number of threads for this process                              
>>>     1
>>>  GLOBAL| This output is from process                                    
>>>    0
>>> also I used 'top' command and found that only 6 cores were busy. 
>>> I use 2 x 24 core processor, and set:
>>> export OMP_NUM_THREADS=8
>>> mpirun -n 6 /lib/CP2K/cp2k/exe/local/cp2k.popt -i project.inp -o 
>>> output.out
>>> Any suggestion will be greatly appreciated..
>>>
>>>
>>>
>>> 在 2019年9月20日星期五 UTC+8下午10:45:55，Pierre Cazade写道： 
>>>>
>>>> Hello Nikhil,
>>>>
>>>> Withe command "mpirun -n 42 cp2k.pop -i inp.inp -o -out.out", you are 
>>>> requesting 42 MPI threads and not 42 OpenMP threads. MPI usually relies on 
>>>> replicated data which means that, for a poorly program software, it will 
>>>> request a total amount of memory which the amount of memory required by a 
>>>> scalar execution times the number of threads. This can very quickly become 
>>>> problematic, in particular for QM calculations. OpenMP, however relies on 
>>>> shared memory, the data is normally not replicated but shared between 
>>>> threads and therefore, in an ideal scenario, the amount of memory needed 
>>>> for 42 OpenMP threads is the same as a single one.
>>>>
>>>> This might explains why you calculation freezes. You are out of memory. 
>>>> On your workstation, you should only use the executable "cp2k.ssmp" which 
>>>> is the OpenMP version. Then you don't need the mpirun command:
>>>>
>>>> cp2k.ssmp -i inp.inp -o -out.out
>>>>
>>>> To control the number of OpenMP threads, set the env variable: 
>>>> OMP_NUM_THREADS, e.g. in bash, export OMP_NUM_THREADS=48
>>>>
>>>> Now, if you need to balance between MPI and OpenMP, you should use the 
>>>> executable named cp2k.psmp. Here is such an example:
>>>>
>>>> export OMP_NUM_THREADS=24
>>>> mpirun -n 2 cp2k.psmp -i inp.inp -o -out.out
>>>>
>>>> In this example, I am requesting two MPI threads and each of them can 
>>>> use up to 24 OpenMP threads.
>>>>
>>>> Hope this clarifies things for you.
>>>>
>>>> Regards,
>>>> Pierre
>>>>
>>>> On 20/09/2019 14:09, Nikhil Maroli wrote:
>>>>
>>>> Dear all, 
>>>>
>>>> I have installed all the versions of CP2K in my workstation with 2 x 12 
>>>> core processor, total thread=48
>>>>
>>>> I wanted to run cp2k in parallel using 42 threads, can anyone share the 
>>>> commands that i can use.
>>>>
>>>> I have tried 
>>>>
>>>> mpirun -n 42 cp2k.pop -i inp.inp -o -out.out
>>>>
>>>> After this command there is a rise in memory to 100 % and the whole 
>>>> system freezes. (i have 128GB ram).
>>>>
>>>> Any suggestion will be greatly appreciated,
>>>> -- 
>>>> You received this message because you are subscribed to the Google 
>>>> Groups "cp2k" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send 
>>>> an email to c... at googlegroups.com.
>>>> To view this discussion on the web visit 
>>>> https://groups.google.com/d/msgid/cp2k/39284c57-f6eb-463e-81a6-3a123596a9f2%40googlegroups.com 
>>>> <https://groups.google.com/d/msgid/cp2k/39284c57-f6eb-463e-81a6-3a123596a9f2%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>> .
>>>>
>>>>
>>>> -- 
>>>> Dr Pierre Cazade, PhD
>>>> AD3-023, Bernal Institute,
>>>> University of Limerick,
>>>> Plassey Park Road,
>>>> Castletroy, co. Limerick,
>>>> Ireland
>>>>
>>>> -- 
>>> You received this message because you are subscribed to the Google 
>>> Groups "cp2k" group.
>>> To unsubscribe from this group and stop receiving emails from it, send 
>>> an email to c... at googlegroups.com.
>>> To view this discussion on the web visit 
>>> https://groups.google.com/d/msgid/cp2k/2d26c264-d73a-44de-af89-13ff0ac86f69%40googlegroups.com 
>>> <https://groups.google.com/d/msgid/cp2k/2d26c264-d73a-44de-af89-13ff0ac86f69%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>>
>>> -- 
>>> You received this message because you are subscribed to the Google 
>>> Groups "cp2k" group.
>>> To unsubscribe from this group and stop receiving emails from it, send 
>>> an email to c... at googlegroups.com.
>>> To view this discussion on the web visit 
>>> https://groups.google.com/d/msgid/cp2k/AM6PR10MB26968C27CCEC14903050632EAE800%40AM6PR10MB2696.EURPRD10.PROD.OUTLOOK.COM 
>>> <https://groups.google.com/d/msgid/cp2k/AM6PR10MB26968C27CCEC14903050632EAE800%40AM6PR10MB2696.EURPRD10.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer>
>>> .
>>>
>>
>>
>> -- 
>>
>> -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
>>   Ari Paavo Seitsonen / Ar... at iki.fi / http://www.iki.fi/~apsi/
>>     Ecole Normale Supérieure (ENS), Département de Chimie, Paris
>>     Mobile (F) : +33 789 37 24 25    (CH) : +41 79 71 90 935
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20190930/736f635c/attachment.htm>