| | | | | | | requested parallelism | walltime sum (s) | nb instances | any sync average per thread time (s) | any wait average per thread time (s) | parallelism overhead (%) | local speedup if perfectly balanced | global speedup if perfectly balanced |
start addr | function name | source location | level | ancestor thread num | invoker | parallel or teams | 1x52 | 1x52 | 1x52 | 1x52 | 1x52 | 1x52 | 1x52 | 1x52 |
libgromacs_mpi.so.7:0x319bbc | nonbonded_verlet_t::dispatchNonbondedKernel(gmx::Interaction... | kerneldispatch.cpp:265 | 0 | 0 | runtime | parallel | 52 | 18.886 | 10.0 E3 | 0.837 | 0.834 | 4.43 | 1.046 | 1.015 |
libgromacs_mpi.so.7:0x874ffb | gmx_pme_do(gmx_pme_t*, gmx::ArrayRef<gmx::BasicVector<float>... | pme.cpp:1296 | 0 | 0 | runtime | parallel | 52 | 8.533 | 10.0 E3 | 1.217 | 1.198 | 14.3 | 1.166 | 1.022 |
libgromacs_mpi.so.7:0x7a8a99 | gmx::constrain_lincs(bool, t_inputrec const&, long, gmx::Lin... | lincs.cpp:2483 | 0 | 0 | runtime | parallel | 52 | 4.149 | 10.0 E3 | 1.768 | 1.712 | 42.6 | 1.743 | 1.033 |
libgromacs_mpi.so.7:0x88f407 | spread_on_grid(gmx_pme_t const*, PmeAtomComm*, pmegrids_t co... | pme_spread.cpp:936 | 0 | 0 | runtime | parallel | 52 | 3.455 | 10.0 E3 | 0.618 | 0.616 | 17.9 | 1.218 | 1.011 |
libgromacs_mpi.so.7:0x88f477 | spread_on_grid(gmx_pme_t const*, PmeAtomComm*, pmegrids_t co... | pme_spread.cpp:1010 | 0 | 0 | runtime | parallel | 52 | 2.704 | 10.0 E3 | 0.295 | 0.294 | 10.9 | 1.123 | 1.005 |
libgromacs_mpi.so.7:0x8750fb | gmx_pme_do(gmx_pme_t*, gmx::ArrayRef<gmx::BasicVector<float>... | pme.cpp:1401 | 0 | 0 | runtime | parallel | 52 | 2.570 | 10.0 E3 | 0.497 | 0.495 | 19.3 | 1.240 | 1.009 |
libgromacs_mpi.so.7:0x85f66f | ListedForces::calculate(gmx_wallcycle*, float const (*) [3],... | listed_forces.cpp:385 | 0 | 0 | runtime | parallel | 52 | 2.447 | 10.0 E3 | 0.644 | 0.642 | 26.3 | 1.357 | 1.012 |
libgromacs_mpi.so.7:0x30a0b8 | reduceForces(nbnxn_atomdata_t*, gmx::AtomLocality, Nbnxm::Gr... | atomdata.cpp:1188 | 0 | 0 | runtime | parallel | 52 | 0.992 | 10.0 E3 | 0.265 | 0.263 | 26.7 | 1.364 | 1.005 |
libgromacs_mpi.so.7:0x38ca3f | PairlistSet::constructPairlists(gmx::InteractionLocality, Nb... | pairlist.cpp:3958 | 0 | 0 | runtime | parallel | 52 | 0.869 | 126 | 92.0 E-3 | 92.0 E-3 | 10.6 | 1.118 | 1.002 |
libgromacs_mpi.so.7:0x88f365 | spread_on_grid(gmx_pme_t const*, PmeAtomComm*, pmegrids_t co... | pme_spread.cpp:910 | 0 | 0 | runtime | parallel | 52 | 0.744 | 10.0 E3 | 0.157 | 0.155 | 21.1 | 1.267 | 1.003 |
libgromacs_mpi.so.7:0x77965e | gmx::Constraints::Impl::apply(bool, bool, long, int, float, ... | constr.cpp:587 | 0 | 0 | runtime | parallel | 52 | 0.706 | 10.0 E3 | 0.205 | 0.204 | 29.1 | 1.410 | 1.004 |
libgromacs_mpi.so.7:0x7d178a | gmx::Update::Impl::update_coords(t_inputrec const&, long, in... | update.cpp:1687 | 0 | 0 | runtime | parallel | 52 | 0.458 | 10.0 E3 | 0.103 | 0.102 | 22.6 | 1.292 | 1.002 |
libgromacs_mpi.so.7:0x309f7c | nbnxn_atomdata_copy_x_to_nbat_x(Nbnxm::GridSet const&, gmx::... | atomdata.cpp:1023 | 0 | 0 | runtime | parallel | 52 | 0.447 | 9.88 E3 | 0.250 | 0.249 | 56.0 | 2.273 | 1.004 |
libgromacs_mpi.so.7:0x41864c | gmx::ThreadedForceBuffer<float [4]>::reduce(gmx::ForceWithSh... | threaded_force_buffer.cpp:170 | 0 | 0 | runtime | parallel | 52 | 0.367 | 10.0 E3 | 86.0 E-3 | 84.4 E-3 | 23.4 | 1.306 | 1.002 |
libgromacs_mpi.so.7:0x39bac9 | nonbonded_verlet_t::dispatchPruneKernelCpu(gmx::InteractionL... | prunekerneldispatch.cpp:68 | 0 | 0 | runtime | parallel | 52 | 0.337 | 876 | 32.3 E-3 | 32.1 E-3 | 9.58 | 1.106 | 1.001 |
libgromacs_mpi.so.7:0x3c6d1a | dd_make_local_constraints(gmx_domdec_t*, int, gmx_mtop_t con... | domdec_constraints.cpp:458 | 0 | 0 | runtime | parallel | 52 | 0.334 | 127 | 0.295 | 0.295 | 88.3 | 8.572 | 1.005 |
libgromacs_mpi.so.7:0x7d19ee | gmx::Update::Impl::finish_update(t_inputrec const&, bool, in... | update.cpp:1637 | 0 | 0 | runtime | parallel | 52 | 0.273 | 10.0 E3 | 67.5 E-3 | 65.9 E-3 | 24.7 | 1.329 | 1.001 |
libgromacs_mpi.so.7:0x30a118 | reduceForces(nbnxn_atomdata_t*, gmx::AtomLocality, Nbnxm::Gr... | atomdata.cpp:1272 | 0 | 0 | runtime | parallel | 52 | 0.170 | 10.0 E3 | 46.4 E-3 | 44.7 E-3 | 27.3 | 1.375 | 1.001 |
libgromacs_mpi.so.7:0x87cb85 | unwrap_periodic_pmegrid(gmx_pme_t*, float*) | pme_grid.cpp:543 | 0 | 0 | runtime | parallel | 52 | 0.138 | 10.0 E3 | 53.3 E-3 | 51.7 E-3 | 38.7 | 1.631 | 1.001 |
libgromacs_mpi.so.7:0x3d1fa3 | dd_make_local_top(gmx_domdec_t const&, gmx_domdec_zones_t co... | localtopology.cpp:845 | 0 | 0 | runtime | parallel | 52 | 0.121 | 127 | 43.2 E-3 | 43.2 E-3 | 35.7 | 1.554 | 1.001 |
libgromacs_mpi.so.7:0x7cbf03 | setupForceOutputs(ForceHelperBuffers*, gmx::ArrayRefWithPadd... | sim_util.cpp:473 | 0 | 0 | runtime | parallel | 52 | 90.9 E-3 | 10.0 E3 | 37.7 E-3 | 36.1 E-3 | 41.4 | 1.707 | 1.001 |
libgromacs_mpi.so.7:0x87cb25 | unwrap_periodic_pmegrid(gmx_pme_t*, float*) | pme_grid.cpp:512 | 0 | 0 | runtime | parallel | 52 | 87.0 E-3 | 10.0 E3 | 39.0 E-3 | 37.1 E-3 | 44.8 | 1.811 | 1.001 |
libgromacs_mpi.so.7:0x7ae1d0 | compute_globals(gmx_global_stat*, t_commrec*, t_inputrec con... | md_support.cpp:113 | 0 | 0 | runtime | parallel | 52 | 48.4 E-3 | 2.00 E3 | 12.0 E-3 | 11.7 E-3 | 24.8 | 1.330 | 1.000 |
libgromacs_mpi.so.7:0x38dacf | PairlistSet::constructPairlists(gmx::InteractionLocality, Nb... | pairlist.cpp:3681 | 0 | 0 | runtime | parallel | 52 | 44.7 E-3 | 126 | 10.6 E-3 | 10.6 E-3 | 23.8 | 1.312 | 1.000 |
libgromacs_mpi.so.7:0x314f58 | Nbnxm::Grid::setCellIndices(int, int, Nbnxm::GridSetData*, g... | grid.cpp:1481 | 0 | 0 | runtime | parallel | 52 | 42.5 E-3 | 127 | 18.0 E-3 | 18.0 E-3 | 42.3 | 1.734 | 1.000 |
libgromacs_mpi.so.7:0x7cc825 | postProcessForces(t_commrec const*, long, t_nrnb*, gmx_wallc... | vec.h:538 | 0 | 0 | runtime | parallel | 52 | 31.1 E-3 | 1.00 E3 | 10.8 E-3 | 10.7 E-3 | 34.8 | 1.534 | 1.000 |
libgromacs_mpi.so.7:0x861791 | setup_bonded_threading(bonded_threading_t*, int, bool, Inter... | manage_threading.cpp:396 | 0 | 0 | runtime | parallel | 52 | 30.3 E-3 | 127 | 6.32 E-3 | 6.29 E-3 | 20.9 | 1.264 | 1.000 |
libgromacs_mpi.so.7:0x777a49 | calc_vir(int, float const (*) [3], float const (*) [3], floa... | calcvir.cpp:109 | 0 | 0 | runtime | parallel | 52 | 20.3 E-3 | 1.00 E3 | 6.02 E-3 | 5.83 E-3 | 29.7 | 1.422 | 1.000 |
libgromacs_mpi.so.7:0x7b0764 | atoms2md(gmx_mtop_t const&, t_inputrec const&, int, gmx::Arr... | mdatoms.cpp:503 | 0 | 0 | runtime | parallel | 52 | 17.5 E-3 | 127 | 3.75 E-3 | 3.72 E-3 | 21.4 | 1.272 | 1.000 |
libgromacs_mpi.so.7:0x417c78 | gmx::ThreadedForceBuffer<float [4]>::ThreadedForceBuffer(int... | threaded_force_buffer.cpp:243 | 0 | 0 | runtime | parallel | 52 | 17.0 E-3 | 1.00 | 6.55 E-3 | 6.55 E-3 | 38.5 | 1.627 | 1.000 |
libgromacs_mpi.so.7:0x7cc249 | setupForceOutputs(ForceHelperBuffers*, gmx::ArrayRefWithPadd... | sim_util.cpp:473 | 0 | 0 | runtime | parallel | 52 | 15.2 E-3 | 1.00 E3 | 3.34 E-3 | 3.19 E-3 | 21.9 | 1.281 | 1.000 |
libgromacs_mpi.so.7:0x7a6e41 | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:1636 | 0 | 0 | runtime | parallel | 52 | 9.45 E-3 | 127 | 1.67 E-3 | 1.64 E-3 | 17.6 | 1.214 | 1.000 |
libgromacs_mpi.so.7:0x7a647c | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:2196 | 0 | 0 | runtime | parallel | 52 | 9.09 E-3 | 127 | 2.33 E-3 | 2.30 E-3 | 25.6 | 1.344 | 1.000 |
libgromacs_mpi.so.7:0x7a7666 | gmx::set_lincs_matrix(gmx::Lincs*, gmx::ArrayRef<float const... | lincs.cpp:1347 | 0 | 0 | runtime | parallel | 52 | 8.93 E-3 | 127 | 2.94 E-3 | 2.83 E-3 | 32.9 | 1.490 | 1.000 |
libgromacs_mpi.so.7:0x3e7325 | dd_redistribute_cg(_IO_FILE*, long, gmx_domdec_t*, int*, t_s... | redistribute.cpp:624 | 0 | 0 | runtime | parallel | 52 | 7.30 E-3 | 126 | 1.94 E-3 | 1.92 E-3 | 26.6 | 1.362 | 1.000 |
libgromacs_mpi.so.7:0x898735 | fft5d_plan_3d(int, int, int, int*, int, t_complex**, t_compl... | fft5d.cpp:643 | 0 | 0 | runtime | parallel | 52 | 7.00 E-3 | 2.00 | 1.11 E-3 | 1.10 E-3 | 15.8 | 1.188 | 1.000 |
libgromacs_mpi.so.7:0x317f15 | Nbnxm::GridSet::putOnGrid(float const (*) [3], int, float co... | gridset.cpp:212 | 0 | 0 | runtime | parallel | 52 | 6.36 E-3 | 127 | 2.50 E-3 | 2.47 E-3 | 39.2 | 1.645 | 1.000 |
libgromacs_mpi.so.7:0x7a70c2 | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:1695 | 0 | 0 | runtime | parallel | 52 | 6.21 E-3 | 127 | 1.53 E-3 | 1.51 E-3 | 24.7 | 1.327 | 1.000 |
libgromacs_mpi.so.7:0x7e0e67 | calc_vcm_grp(t_mdatoms const&, gmx::ArrayRef<gmx::BasicVecto... | vcm.cpp:222 | 0 | 0 | runtime | parallel | 52 | 3.08 E-3 | 102 | 945 E-6 | 871 E-6 | 30.7 | 1.442 | 1.000 |
libgromacs_mpi.so.7:0x7e1b18 | process_and_stopcm_grp(_IO_FILE*, t_vcm*, t_mdatoms const&, ... | vcm.cpp:374 | 0 | 0 | runtime | parallel | 52 | 1.60 E-3 | 102 | 802 E-6 | 735 E-6 | 50.1 | 2.005 | 1.000 |
libgromacs_mpi.so.7:0x8988df | fft5d_plan_3d(int, int, int, int*, int, t_complex**, t_compl... | fft5d.cpp:643 | 0 | 0 | runtime | parallel | 52 | 1.01 E-3 | 2.00 | 456 E-6 | 455 E-6 | 45.0 | 1.818 | 1.000 |
libgromacs_mpi.so.7:0x89880b | fft5d_plan_3d(int, int, int, int*, int, t_complex**, t_compl... | fft5d.cpp:643 | 0 | 0 | runtime | parallel | 52 | 919 E-6 | 2.00 | 434 E-6 | 433 E-6 | 47.2 | 1.894 | 1.000 |
libgromacs_mpi.so.7:0x886d3e | pme_init_all_work(pme_solve_work_t**, int, int) | pme_solve.cpp:153 | 0 | 0 | runtime | parallel | 52 | 31.2 E-6 | 1.00 | 7.43 E-6 | 7.29 E-6 | 23.8 | 1.312 | 1.000 |
libgromacs_mpi.so.7:0x4036aa | gmx_ekindata_t::gmx_ekindata_t(int, float, int) | group.cpp:66 | 0 | 0 | runtime | parallel | 52 | 9.04 E-6 | 1.00 | 3.36 E-6 | 3.18 E-6 | 37.2 | 1.592 | 1.000 |
libgromacs_mpi.so.7:0x877f0b | PmeAtomComm& std::vector<PmeAtomComm, std::allocator<PmeAtom... | vector.tcc:480 | 0 | 0 | runtime | parallel | 52 | 9.00 E-6 | 1.00 | 2.52 E-6 | 2.35 E-6 | 28.1 | 1.390 | 1.000 |