Andreas Lauser
a6191a2345
the new home of Valgrind.hpp is opm-common!
2016-12-14 12:38:12 +01:00
Andreas Lauser
fb8d984bf1
adapt to the move of quad.hh to opm-material
2016-12-14 12:38:12 +01:00
Andreas Lauser
e1d11da6d6
EclProblem: fix the type returned by the const version of materialLawManager()
...
it is supposed to a shared pointer to a constant object, not a
constant pointer to a mutable object...
2016-12-10 10:53:28 +01:00
Andreas Lauser
4b2156a071
ebos: update the PffGridVector for transmissibility after geologic events
2016-12-08 12:26:32 +01:00
Andreas Lauser
706e51f878
ebos: store the names wells that are disabled on the local process due to load balancing
2016-12-07 20:24:48 +01:00
Andreas Lauser
5067ce2f27
ebos: use the transmissibilities as edge weights for load balancing
...
this makes creating the grid a bit slower because the
transmissibilities need to be calculated twice: once for the
sequential grid and once for the distributed one. while corresponds to
the way `flow_legacy` does the load balancing and it should allow
better results, this does not seem to be the case for the Norne deck
if ZOLTAN is not available:
After loadbalancing process 3 has 4413 cells.
After loadbalancing process 2 has 12390 cells.
After loadbalancing process 0 has 13629 cells.
After loadbalancing process 1 has 21253 cells.
i.e., process 1 is responsible for almost 5 as many cells as process
3.
2016-12-07 20:24:48 +01:00
Andreas Lauser
a0481039db
ebos: add the disabled well names kluge to the grid manager
...
this array is only non-empty for parallel runs.
2016-12-07 20:24:48 +01:00
Andreas Lauser
36c010c956
ebos: replace the opm-parser related pointers/smart pointers by references
...
this seems to be the spirit of the season. note that now the deck
object must be copied, but I suppose that copying it is pretty cheap.
2016-12-07 20:24:48 +01:00
Andreas Lauser
9460177f17
CollectDataToIoRank: don't require the elements of a dune grid to be traversed in consecutive order
2016-12-05 13:56:32 +01:00
Robert Kloefkorn
4921f7d817
When using matrialLawManager then elemIdx has to be used.
2016-12-05 13:47:04 +01:00
Robert Kloefkorn
e3286b3ec0
EclEquil: use std::vector to store mappings.
2016-12-02 18:55:27 +01:00
Robert Kloefkorn
e240cf8eae
[bugfix] fix equil and grid mappings in initialization.
2016-12-02 18:54:14 +01:00
Robert Kloefkorn
deed854a20
[bugfix] fix global, distributed and local mappings.
2016-12-02 18:51:19 +01:00
Robert Kloefkorn
0062a0e18f
[bugfix] use neighbor to check whether neighbor is present.
2016-12-02 18:51:19 +01:00
Jørgen Kvalsvik
3088068e37
Mini deck is reference, not pointer
2016-12-01 16:17:52 +01:00
Andreas Lauser
4a0832030c
EclProblem: add docstrings for the materialLawManager() methods
2016-11-30 14:24:15 +01:00
Andreas Lauser
4ef6d5d1ce
EclProblem: add a porosity(elemIdx) method
...
this is not a generic API but makes emulating legacy stuff which
requires the porosity quite a bit easier in flow_ebos.
2016-11-30 14:21:04 +01:00
Andreas Lauser
a3ddd2066b
ebos: make it possible to easily disable SWATINIT
...
in flow_ebos, this code collides with the flow part.
2016-11-30 14:20:44 +01:00
Andreas Lauser
c4c00bdaaa
implement two-phase blackoil simulations
2016-11-22 15:22:41 +01:00
Andreas Lauser
6cf1eab4f4
ebos: export the ECL material law manager
...
this allows external code (i.e., flow_ebos) to reuse it.
2016-11-22 14:08:07 +01:00
Andreas Lauser
431cebaa21
EclFluxModule: make sure that the upwind and downwind directions are set
...
before, they were undefined for phases which exhibited zero mobility
in both directions. since in this case the flux is zero (and thus the
upstream direction does not matter), the correctness is unaffected by
this patch. (still, I consider it to bug and valgrind also complained
about it.)
2016-11-17 19:48:07 +01:00
Andreas Lauser
a5e859e4c6
discretizations: no longer depend on the problem to provide the mappers
...
Dune entity mappers seem to be dirt-cheap to create so there is not
much point in jumping through hoops to avoid this...
2016-11-17 19:44:17 +01:00
Andreas Lauser
759c2dbdaa
move all applications into their top-level directory
...
thanks to [at]akva2 for the suggestion.
2016-11-11 15:04:04 +01:00
