[aspect-devel] Aspect-devel Digest, Vol 77, Issue 23

[E.M. Dragowsky]

O
​K -- I found the previous install of aspect *did* run on the head nodes of
the cluster, but not on the compute nodes.

I have now rebuilt the debug version, on a compute node, and verified that
the executable executes. make test fails in the usual fully consistent
fashion, because it cannot find the source code for the tests:

​CMake Error: The source directory
"/home/mrd20/src/rider/aspect/build/tests.additional_outputs_02" does not
exist.

To, the more recent questions, here is the output from the solcx benchmark:
-----------------------------------------------------------------------------
-- This is ASPECT, the Advanced Solver for Problems in Earth's ConvecTion.
--     . version 1.5.0
--     . running in DEBUG mode
--     . running with 1 MPI process
--     . using Trilinos
-----------------------------------------------------------------------------

Loading shared library <./libsolcx.so>

Number of active cells: 256 (on 5 levels)
Number of degrees of freedom: 3,556 (2,178+289+1,089)

*** Timestep 0:  t=0 years
   Rebuilding Stokes preconditioner...
   Solving Stokes system... 30+4 iterations.
      Relative Stokes residual after nonlinear iteration 1: 1
   Solving Stokes system... 0+0 iterations.
      Relative Stokes residual after nonlinear iteration 2: 1.74731e-08

   Postprocessing:
     Errors u_L1, p_L1, u_L2, p_L2: 1.125998e-06, 2.994139e-03,
1.670011e-06, 9.778440e-03
     Writing graphical output:      output/solution/solution-00000

Termination requested by criterion: end time


+---------------------------------------------+------------+------------+
| Total wallclock time elapsed since start    |      1.03s |            |
|                                             |            |            |
| Section                         | no. calls |  wall time | % of total |
+---------------------------------+-----------+------------+------------+
| Assemble Stokes system          |         2 |    0.0318s |       3.1% |
| Build Stokes preconditioner     |         1 |     0.221s |        22% |
| Solve Stokes system             |         2 |    0.0613s |         6% |
| Initialization                  |         1 |     0.207s |        20% |
| Postprocessing                  |         1 |     0.158s |        15% |
| Setup dof systems               |         1 |     0.104s |        10% |
| Setup initial conditions        |         1 |    0.0466s |       4.5% |
+---------------------------------+-----------+------------+------------+


And attached is the contents of tests.

Thanks for your patience
~ Em


On Wed, Apr 25, 2018 at 5:05 PM, E.M. Dragowsky <dragowsky at case.edu> wrote:

> so you are saying the copy of the 'tests' code should occur during 'make
> setup_tests'?
>
> yes, there is definitely an issue, as ./aspect does not 'run'.
> which is frustrating, as the first time around it ran fine.
>
> I'll review again my notes on how I built the first 'debug' version, and
> try again from scratch.
>
> On Wed, Apr 25, 2018 at 4:56 PM, Timo Heister <heister at clemson.edu> wrote:
>
>> >> 1. Can you show the output of "ls -al ." inside the "tests" folder
>> >> inside your "build" folder?
>> >>
>> > I've attached a file. Performing the list revealed that the tests
>> themselves were not present, just the 'output' files anticipating the test
>> output. So then from ~/aspect/build,  I copied the contents of ../tests to
>> ~/aspect/build/tests.
>>
>> No no no, don't copy anything around. This will all be done
>> automatically as needed.
>>
>> > Nothing happends when I try to run aspect against a test directly. An
>> aspect process is launched, but it just sleeps.
>> > I've tried also against ~/aspect/benchmarks/solcx.
>> > Again, an aspect process launches, but just sleeps.
>>
>> Then that is the issue. What happens if you just run "./aspect"?
>>
>>
>> --
>> Timo Heister
>> http://www.math.clemson.edu/~heister/
>> _______________________________________________
>> Aspect-devel mailing list
>> Aspect-devel at geodynamics.org
>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel
>>
>
>
>
> --
> E.M. Dragowsky, Ph.D.
> Research Computing -- UTech
> Case Western Reserve University
> (216) 368-0082
>



-- 
E.M. Dragowsky, Ph.D.
Research Computing -- UTech
Case Western Reserve University
(216) 368-0082
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.geodynamics.org/pipermail/aspect-devel/attachments/20180426/f8719c74/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: build-tests-ls-laR.26apr
Type: application/octet-stream
Size: 935971 bytes
Desc: not available
URL: <http://lists.geodynamics.org/pipermail/aspect-devel/attachments/20180426/f8719c74/attachment-0001.obj>