Hi,
From the 'relax -i' printout, I can see that you have the Intel
i7-4770 CPU. This has 4 cores and hyperthreading. I can also see
that mpi4py is installed. Therefore you can run relax about 7 times
faster! You should then run:
$ cd my_relax_results_directory
$ mpirun -np 8 /usr/local/bin/relax --multi='mpi4py' --gui --tee analysis.log
For details about Gary Thompson's multi-processor framework, see the
introduction chapter of the manual
(http://www.nmr-relax.com/manual/multi_processor_framework.html).
Note the single and double '-' usage here, that makes a big
difference.
From the log messages you attached
(https://gna.org/bugs/download.php?file_id=22465), I can see that
something just stopped. The last of the messages are:
"""
Storing the optimisation results for the spin
'#BpUreE_Nterm_mol1:70@N', the optimised chi-squared value is lower
than the current value (179.19383555 < None).
Fitting to spin '#BpUreE_Nterm_mol1:71@N'
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Method of Multipliers
~~~~~~~~~~~~~~~~~~~~~
k: 0 xk: [ 12000, 0.6, 0.6,
0] fk: 73866.4675956
Entering sub-algorithm.
Newton minimisation
~~~~~~~~~~~~~~~~~~~
Line search: Backtracking line search.
Hessian modification: The Gill, Murray, and Wright modified
Cholesky algorithm.
k: 0 xk: [ 12000, 0.6,
0.6, 0] fk: 73866.4675956
Parameter values: [24397.172656690935, 1.0404007478636053,
0.98101345987882127, 30.269140680559676]
Function value: -506.91372361949527
Iterations: 8
Function calls: 21
Gradient calls: 9
Hessian calls: 8
Warning: None
"""
Are you sure that this what you see at the end of the relax controller
window? This is not right, relax cannot just stop here. This is in
the middle of executing the minfx optimisation library
(https://gna.org/projects/minfx/). Is something still happening in
the relax controller? Is it still at this point?
Note that the model-free auto-analysis is designed to continue from
where it stopped. Therefore if you run relax again with the same set
up and the same results directory specified, it will load the results
already there and continue from that point.
Cheers,
Edward
On 30 September 2014 14:59, anonymous <NO-REPLY.INVALID-ADDRESS@xxxxxxx>
wrote:
Follow-up Comment #4, bug #22730 (project relax):
Can I keep writing here?
I have organized a separate folder for the output results, as you
recommended.
Started relax as 'relax_gui -log log'. The first messages from the runnung
relax controller I copied and attached in the file. These are exactly those
that I recieved running it earlier. Also, typed 'relax -i' and the result is
attached as well.
When I noticed the interruption of the calculations back then, the Relax
windows were disappeared, perhaps somehow automatically.
Could you see anything now from these files?
Olena
(file #22464, file #22465)
_______________________________________________________
Additional Item Attachment:
File name: Relax -i Size:5 KB
File name: From Controller.tar.gz Size:228 KB
_______________________________________________________
Reply to this item at:
<http://gna.org/bugs/?22730>
_______________________________________________
Message sent via/by Gna!
http://gna.org/
Re: [bug #22730] Model-free auto-analysis - relax stops and quits at the polate step.