71 Commits

Author SHA1 Message Date
Anthony Berg
e7cd6ae34a feat: add progress bar for Autotuner 2025-03-26 00:28:16 +01:00
Anthony Berg
c485b37219 fix(GPUSimulators): swap out missing functions in Autotuner 2025-02-18 10:22:00 +01:00
Anthony Berg
aa54d21f0a feat(GPUSimulator): add the extra variables introduced in SWWECommon for all other algorithms 2025-02-18 10:21:12 +01:00
Anthony Berg
71cc5628c8 fix(GPUSimulator): use class function instead of calling hash_kernel separately 2025-02-18 10:16:10 +01:00
Anthony Berg
7c312c255f docs(GPUSimulators): add documentation for CudaContext 2025-02-14 16:29:47 +01:00
Anthony Berg
ef207432db refactor(GPUSimulator): follow PEP 8 style guide 2025-02-14 12:40:31 +01:00
Martin Lilleeng Sætra
55029275e3 Cleanup 2022-08-16 19:01:24 +02:00
Martin Lilleeng Sætra
234c8cb727 Merge branch 'master' of github.com:setmar/ShallowWaterGPU 2022-08-16 14:35:21 +00:00
Martin Lilleeng Sætra
403146cccc Draft of SHMEMSimulator 2022-08-16 14:34:57 +00:00
Martin Lilleeng Sætra
a260230002 Removed nvtx and MPI options for job script 2022-05-25 14:52:44 +02:00
Martin Lilleeng Sætra
a588948e77 Bugfix for ghost cell exchange and nc atts 2022-05-25 11:25:10 +00:00
Martin Lilleeng Sætra
e5a8486501 Periodic BC now working with domain decomposition again. 2022-05-09 13:50:59 +00:00
Martin Lilleeng Sætra
c7e6f17445 Adjusting order of substep ops and benchmark scripts 2022-04-26 15:39:21 +02:00
Martin Lilleeng Sætra
acb7d2ab39 Async mem ops 2022-04-26 12:10:22 +00:00
Martin Lilleeng Sætra
6c8bac6f7b Async mem ops 2022-04-26 11:34:29 +00:00
Martin Lilleeng Sætra
dcef56a1b9 More time counters and separation of experiments. 2022-04-11 13:57:33 +00:00
Martin Lilleeng Sætra
d83d620512 Profiling code for DGX-2. Added fixed dt. 2022-04-08 15:09:09 +02:00
Martin Lilleeng Sætra
8fd9d3d850 Added more profiling timers 2022-04-07 09:07:20 +00:00
Martin Lilleeng Sætra
124e033ff5 More profiling 2022-04-01 13:42:56 +00:00
Martin Lilleeng Sætra
11223f0890 Checkpoint. Next: Compose domain before writing 2021-05-23 18:26:18 +00:00
Martin Lilleeng Sætra
ad2be4b5a5 Added sync, some comments, and done some tinkering 2021-05-18 21:19:07 +00:00
Martin Lilleeng Sætra
b6424b84fb First running version. 2021-05-13 15:25:34 +00:00
Martin Lilleeng Sætra
5b6c4f682d Checkpoint SHMEMSimulatorGroup. Needs debugging. 2021-05-12 16:09:29 +00:00
Martin Lilleeng Sætra
2da4640840 Added class for managing SHMEMSimulators 2021-05-05 15:24:34 +00:00
Martin Lilleeng Sætra
98aff206dc Running multiple CUDA contexts per process/thread 2021-02-21 22:57:50 +01:00
Martin Lilleeng Sætra
239c1fdcd9 Refactoring. 2019-08-16 15:27:46 +00:00
Martin Lilleeng Sætra
6a1f58c8df
Minor Linux compatibility fix. 2019-08-12 14:43:11 +00:00
Martin Lilleeng Sætra
75678c4fb3
Made MPI magic Linux compat. 2019-08-06 11:18:04 +00:00
Martin Lilleeng Sætra
d605b9e102
Small bugfixes/updates. 2019-08-05 11:59:38 +00:00
Martin Lilleeng Sætra
4ca9581b37
EulerTesting.ipynb now runs on Linux. Simulation results needs further validation. 2019-06-05 15:30:35 +00:00
André R. Brodtkorb
4fa14abdff Fix for Abel 2018-12-10 16:10:52 +01:00
André R. Brodtkorb
2c6ecc8d88 Pinned memory 2018-12-10 12:08:10 +01:00
André R. Brodtkorb
12174b39db Added example script 2018-12-04 18:06:45 +01:00
André R. Brodtkorb
4292513c03 Fixed global boundary conditions 2018-12-04 14:36:48 +01:00
André R. Brodtkorb
abcda741ab Updated initial conditions 2018-12-03 15:50:05 +01:00
André R. Brodtkorb
ae6404f05e Refactoring 2018-12-03 14:34:28 +01:00
André R. Brodtkorb
b266567d09 Refactoring / cleanup 2018-11-30 11:24:36 +01:00
André R. Brodtkorb
f9f0f20df8 Fixed multigpu for euler 2018-11-29 14:33:29 +01:00
André R. Brodtkorb
b03afc3d81 Fixed general MPI framework 2018-11-28 15:28:35 +01:00
André R. Brodtkorb
6b7de7b755 Bugfix common download / upload 2018-11-21 12:03:52 +01:00
André R. Brodtkorb
c51afef9fc MPI prototype 2018-11-21 07:49:39 +01:00
André R. Brodtkorb
074e38de84 Added mpi utilities 2018-11-20 08:59:42 +01:00
André R. Brodtkorb
cfcaa65bbe Bugfix KP07 and refactoring 2018-11-19 14:07:06 +01:00
André R. Brodtkorb
815b4493b5 Removed fixed timestep size 2018-11-15 17:25:01 +01:00
André R. Brodtkorb
ddac53271c Implemented variable timestep 2018-11-15 16:48:47 +01:00
André R. Brodtkorb
7592ad5b9f Fixed order again 2018-11-15 16:47:13 +01:00
André R. Brodtkorb
dcb849b705 Bugfixes 2018-11-11 15:12:56 +01:00
André R. Brodtkorb
e98ae0a979 Refactoring - broke 2nd order 2018-11-09 11:46:34 +01:00
André R. Brodtkorb
812ebcc9ba Reproducibility 2018-11-08 23:10:35 +01:00
André R. Brodtkorb
77dc93fd3c Added helper files 2018-11-08 22:13:37 +01:00