Commit Graph

100 Commits

Author SHA1 Message Date
Maximilian Gaul
696e6da01d Implement Neighbour list AoS memory layout + performance measurement 2022-01-31 20:27:59 +01:00
Maximilian Gaul
b2a6574426 Remove unnecessary atom force backcopy in computeForce 2022-01-24 18:09:27 +01:00
Maximilian Gaul
c4080e866e Make integrate kernels aware of neighbour list update 2022-01-24 18:04:50 +01:00
Maximilian Gaul
dc4d5f1a9c Porting atom velocity memory layout to AoS, porting velocity integration to CUDA, adding measurements + logbook update 2022-01-01 18:18:12 +01:00
Maximilian Gaul
50007216ed Implemented atom force AoS memory layout, added performance measurements + logbook Update 2022-01-01 16:09:21 +01:00
Maximilian Gaul
72e4599acc Copy neighbour lists only when reneighbouring happens, added measurements + logbook update 2022-01-01 12:56:42 +01:00
Maximilian Gaul
8fa03733e9 Copy parameters & cutforces threshold only once at the start + measurements 2021-12-28 16:48:26 +01:00
Maximilian Gaul
bf1ae3d013 Removed debug prints, only zero atom forces and not copy them, added measurements 2021-12-28 16:32:54 +01:00
Maximilian Gaul
8009b54113 Trying to debug segfault if cudaMemcpy is limited to neighbour list update 2021-12-25 15:36:08 +01:00
Maximilian Gaul
0ea0587442 Only malloc once at the beginning plus measurement csv 2021-12-25 13:52:33 +01:00
Maximilian Gaul
134e3f4b78 Also pinnend neighbor-struct memory, added additional performance measurements, added nvprof result to logbook 2021-12-18 15:58:56 +01:00
Maximilian Gaul
c2bfa3ca3f Add scripts for perf measurement, made atom-memory allocation pinnend using 'cudaMallocHost', added measurements for atom pinnend memory 2021-12-18 13:02:04 +01:00
Maximilian Gaul
2a099da5b7 Started cuda profiling, added first result to logbook 2021-12-03 08:13:43 +01:00
Maximilian Gaul
7691b23d67 Measure memory transfer of CPU to GPU, add explanation how to distribute calculation among multiple GPUs 2021-12-01 17:16:32 +01:00
Maximilian Gaul
da90466f98 Added first performance measurements with threads per block from 1 to 32 2021-11-25 08:09:20 +01:00
Maximilian Gaul
8f723c1299 Added command line description of MD-Bench, added memory transfer rate from CPU to GPU to force.cu 2021-11-23 15:55:23 +01:00
Maximilian Gaul
0586ef150a Fix num of threads instead of num of blocks, add logbook template 2021-11-15 19:39:09 +01:00
Maximilian Gaul
2e5d973f7d Rough rewrite to execute outer loop of force calculation in parallel, not inner loop 2021-11-14 10:02:23 +01:00
Maximilian Gaul
e2fd1a0476 Fixed bug, results are now equal to master branch (but still slow) 2021-11-11 21:00:30 +01:00
Maximilian Gaul
4105c844c6 Runs fine (but slow), results seem to be slightly off from original 2021-11-11 20:47:06 +01:00
Maximilian Gaul
1f5c9c4b23 Fixed segfault error, added more cudaErrorChecks, added cudaFree to avoid memory leak 2021-11-11 20:29:14 +01:00
Maximilian Gaul
29e115464b Fixed cudaMemcpy for AOS data layout, added debug outputs, added cudaErrorChecks 2021-11-11 20:14:30 +01:00
Maximilian Gaul
1a54314c8b First run but segfault at the moment after a few seconds 2021-11-11 15:23:46 +01:00
Maximilian Gaul
280f595b7f Fixed linker error by putting includes and cuda function in extern 'C' 2021-11-11 14:49:29 +01:00
Maximilian Gaul
3428974730 getTimeStamp() couldn't get linked 2021-11-11 08:03:56 +01:00
Maximilian Gaul
b54842f764 Added Makefile instructions for .cu files 2021-11-11 07:27:12 +01:00
Maximilian Gaul
9730164e6f Rename force.c to force.cu because of cuda build errors 2021-11-10 16:20:04 +01:00
Maximilian Gaul
0f5fdd3708 Sum results after cuda function executed 2021-11-10 16:02:05 +01:00
Maximilian Gaul
3f7fb7f22a cudaMemcpy of Atom and other properties, first draft implementation of CUDA kernel 2021-11-09 16:40:25 +01:00
Maximilian Gaul
bfa6c581c3 Copy necessary values for force calculation into cuda memory 2021-11-09 08:37:37 +01:00
Maximilian Gaul
fd886e77eb Added make config for NVCC 2021-11-08 20:32:12 +01:00
Rafael Ravedutti
ec556eb117 Add EAM without explicit types and update fp for PBC atoms
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-11-03 00:57:24 +01:00
Rafael Ravedutti
0f1e824507 Fix integration with mass and add VTK output
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-29 16:52:19 +02:00
Rafael Ravedutti
d9c9cba12d Set EAM parameters for same case as miniMD
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-27 01:38:56 +02:00
Rafael Ravedutti
70cc6aeb19 Add first working version of EAM
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-26 13:55:14 +02:00
Rafael Ravedutti
7db14b2ffe Merge branch 'master' of github.com:RRZE-HPC/MD-Bench 2021-10-26 12:34:43 +02:00
Rafael Ravedutti
d7056f09f4 Fix illegal accesses due to invalid type of ghost atoms
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-26 12:34:35 +02:00
Jan Eitzinger
70e24d7c2c Merge branch 'master' of github.com:RRZE-HPC/MD-Bench 2021-10-26 09:16:31 +02:00
Jan Eitzinger
3c3d27b48a Introduce separate version for traced force routine. 2021-10-26 09:11:17 +02:00
Rafael Ravedutti
99d6a4bdd8 Fix Funcfl reference to avoid segfaults
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-26 01:40:02 +02:00
Rafael Ravedutti
40ddc9ad50 Fix errors introduced by last changes
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-26 01:19:11 +02:00
Rafael Ravedutti
2dac10469c Add EAM force field
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-26 00:40:39 +02:00
Rafael Ravedutti
9d16bb46c8 Include average neighbors and SIMD iterations per atom on stats
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-20 22:43:08 +02:00
Rafael Ravedutti
557c5e4d3f IUpdate logbook again and do not check for variant on ATOMS_LOOP_RUNS effect
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-14 00:46:30 +02:00
Rafael Ravedutti
40bbaac078 Add hardware information on logbook
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-13 22:27:34 +02:00
Rafael Ravedutti
43ba28e130 Add ATOMS_LOOP_RUNS option and statistics to stub variant
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-12 22:39:54 +02:00
Rafael Ravedutti
55d346510e Compute statistics, useful data volume and cycles per SIMD iteration
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-12 15:04:08 +02:00
Rafael Ravedutti
f295f54fca Add useful data volume calculation for force kernel
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-11 16:57:02 +02:00
Rafael Ravedutti
b1f8e91f0f Add cache measurements for versions with and without force computation
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-04 18:13:17 +02:00
Rafael Ravedutti
749c4d55ab Provide trace output for all reneighboring steps
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-04 16:47:07 +02:00