| | | | | | | 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 | 21.395 | 10.0 E3 | 2.971 | 2.969 | 13.9 | 1.161 | 1.040 |
libgromacs_mpi.so.7:0x874ffb | gmx_pme_do(gmx_pme_t*, gmx::ArrayRef<gmx::BasicVector<float>... | pme.cpp:1296 | 0 | 0 | runtime | parallel | 52 | 12.461 | 10.0 E3 | 2.927 | 2.908 | 23.5 | 1.307 | 1.040 |
libgromacs_mpi.so.7:0x7a8a99 | gmx::constrain_lincs(bool, t_inputrec const&, long, gmx::Lin... | lincs.cpp:2483 | 0 | 0 | runtime | parallel | 52 | 6.713 | 10.0 E3 | 3.557 | 3.496 | 53.0 | 2.127 | 1.049 |
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 | 5.637 | 10.0 E3 | 1.669 | 1.667 | 29.6 | 1.421 | 1.022 |
libgromacs_mpi.so.7:0x8750fb | gmx_pme_do(gmx_pme_t*, gmx::ArrayRef<gmx::BasicVector<float>... | pme.cpp:1401 | 0 | 0 | runtime | parallel | 52 | 5.388 | 10.0 E3 | 2.088 | 2.086 | 38.7 | 1.633 | 1.028 |
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 | 5.155 | 10.0 E3 | 1.153 | 1.152 | 22.4 | 1.288 | 1.015 |
libgromacs_mpi.so.7:0x85f66f | ListedForces::calculate(gmx_wallcycle*, float const (*) [3],... | listed_forces.cpp:385 | 0 | 0 | runtime | parallel | 52 | 2.832 | 10.0 E3 | 0.838 | 0.836 | 29.6 | 1.420 | 1.011 |
libgromacs_mpi.so.7:0x30a0b8 | reduceForces(nbnxn_atomdata_t*, gmx::AtomLocality, Nbnxm::Gr... | atomdata.cpp:1188 | 0 | 0 | runtime | parallel | 52 | 1.666 | 10.0 E3 | 0.322 | 0.320 | 19.3 | 1.240 | 1.004 |
libgromacs_mpi.so.7:0x77965e | gmx::Constraints::Impl::apply(bool, bool, long, int, float, ... | constr.cpp:587 | 0 | 0 | runtime | parallel | 52 | 1.120 | 10.0 E3 | 0.367 | 0.366 | 32.8 | 1.488 | 1.005 |
libgromacs_mpi.so.7:0x38ca3f | PairlistSet::constructPairlists(gmx::InteractionLocality, Nb... | pairlist.cpp:3958 | 0 | 0 | runtime | parallel | 52 | 0.945 | 126 | 0.167 | 0.167 | 17.7 | 1.215 | 1.002 |
libgromacs_mpi.so.7:0x7d178a | gmx::Update::Impl::update_coords(t_inputrec const&, long, in... | update.cpp:1687 | 0 | 0 | runtime | parallel | 52 | 0.934 | 10.0 E3 | 0.322 | 0.320 | 34.5 | 1.527 | 1.004 |
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 | 0.880 | 10.0 E3 | 0.244 | 0.242 | 27.7 | 1.383 | 1.003 |
libgromacs_mpi.so.7:0x41864c | gmx::ThreadedForceBuffer<float [4]>::reduce(gmx::ForceWithSh... | threaded_force_buffer.cpp:170 | 0 | 0 | runtime | parallel | 52 | 0.762 | 10.0 E3 | 0.243 | 0.242 | 31.9 | 1.469 | 1.003 |
libgromacs_mpi.so.7:0x309f7c | nbnxn_atomdata_copy_x_to_nbat_x(Nbnxm::GridSet const&, gmx::... | atomdata.cpp:1023 | 0 | 0 | runtime | parallel | 52 | 0.659 | 9.88 E3 | 0.330 | 0.329 | 50.2 | 2.007 | 1.004 |
libgromacs_mpi.so.7:0x7d19ee | gmx::Update::Impl::finish_update(t_inputrec const&, bool, in... | update.cpp:1637 | 0 | 0 | runtime | parallel | 52 | 0.479 | 10.0 E3 | 0.180 | 0.179 | 37.7 | 1.604 | 1.002 |
libgromacs_mpi.so.7:0x39bac9 | nonbonded_verlet_t::dispatchPruneKernelCpu(gmx::InteractionL... | prunekerneldispatch.cpp:68 | 0 | 0 | runtime | parallel | 52 | 0.404 | 876 | 79.4 E-3 | 79.2 E-3 | 19.7 | 1.245 | 1.001 |
libgromacs_mpi.so.7:0x87cb85 | unwrap_periodic_pmegrid(gmx_pme_t*, float*) | pme_grid.cpp:543 | 0 | 0 | runtime | parallel | 52 | 0.337 | 10.0 E3 | 0.156 | 0.154 | 46.3 | 1.861 | 1.002 |
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 | 0.335 | 127 | 0.286 | 0.286 | 85.2 | 6.736 | 1.004 |
libgromacs_mpi.so.7:0x30a118 | reduceForces(nbnxn_atomdata_t*, gmx::AtomLocality, Nbnxm::Gr... | atomdata.cpp:1272 | 0 | 0 | runtime | parallel | 52 | 0.283 | 10.0 E3 | 0.145 | 0.143 | 51.1 | 2.047 | 1.002 |
libgromacs_mpi.so.7:0x87cb25 | unwrap_periodic_pmegrid(gmx_pme_t*, float*) | pme_grid.cpp:512 | 0 | 0 | runtime | parallel | 52 | 0.200 | 10.0 E3 | 0.132 | 0.129 | 65.7 | 2.916 | 1.002 |
libgromacs_mpi.so.7:0x7cbf03 | setupForceOutputs(ForceHelperBuffers*, gmx::ArrayRefWithPadd... | sim_util.cpp:473 | 0 | 0 | runtime | parallel | 52 | 0.187 | 10.0 E3 | 0.124 | 0.122 | 66.3 | 2.968 | 1.002 |
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 | 0.122 | 127 | 45.3 E-3 | 45.2 E-3 | 37.2 | 1.593 | 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 | 79.3 E-3 | 2.00 E3 | 34.5 E-3 | 34.1 E-3 | 43.5 | 1.770 | 1.000 |
libgromacs_mpi.so.7:0x38dacf | PairlistSet::constructPairlists(gmx::InteractionLocality, Nb... | pairlist.cpp:3681 | 0 | 0 | runtime | parallel | 52 | 56.6 E-3 | 126 | 20.9 E-3 | 20.9 E-3 | 37.0 | 1.587 | 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 | 51.0 E-3 | 1.00 E3 | 24.6 E-3 | 24.4 E-3 | 48.2 | 1.932 | 1.000 |
libgromacs_mpi.so.7:0x314f58 | Nbnxm::Grid::setCellIndices(int, int, Nbnxm::GridSetData*, g... | grid.cpp:1481 | 0 | 0 | runtime | parallel | 52 | 48.3 E-3 | 127 | 21.2 E-3 | 21.2 E-3 | 43.8 | 1.780 | 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 | 46.9 E-3 | 127 | 21.3 E-3 | 21.3 E-3 | 45.5 | 1.834 | 1.000 |
libgromacs_mpi.so.7:0x7cc825 | postProcessForces(t_commrec const*, long, t_nrnb*, gmx_wallc... | vec.h:538 | 0 | 0 | runtime | parallel | 52 | 45.4 E-3 | 1.00 E3 | 19.2 E-3 | 19.0 E-3 | 42.3 | 1.733 | 1.000 |
libgromacs_mpi.so.7:0x7cc249 | setupForceOutputs(ForceHelperBuffers*, gmx::ArrayRefWithPadd... | sim_util.cpp:473 | 0 | 0 | runtime | parallel | 52 | 26.2 E-3 | 1.00 E3 | 12.7 E-3 | 12.6 E-3 | 48.6 | 1.946 | 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 | 23.8 E-3 | 127 | 8.33 E-3 | 8.31 E-3 | 35.0 | 1.539 | 1.000 |
libgromacs_mpi.so.7:0x7a647c | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:2196 | 0 | 0 | runtime | parallel | 52 | 19.0 E-3 | 127 | 11.5 E-3 | 11.5 E-3 | 60.6 | 2.539 | 1.000 |
libgromacs_mpi.so.7:0x417c78 | gmx::ThreadedForceBuffer<float [4]>::ThreadedForceBuffer(int... | threaded_force_buffer.cpp:243 | 0 | 0 | runtime | parallel | 52 | 16.2 E-3 | 1.00 | 6.63 E-3 | 6.63 E-3 | 40.8 | 1.689 | 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 | 13.2 E-3 | 127 | 6.32 E-3 | 6.20 E-3 | 47.8 | 1.915 | 1.000 |
libgromacs_mpi.so.7:0x7a6e41 | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:1636 | 0 | 0 | runtime | parallel | 52 | 12.0 E-3 | 127 | 4.23 E-3 | 4.21 E-3 | 35.3 | 1.545 | 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 | 10.7 E-3 | 2.00 | 1.91 E-3 | 1.91 E-3 | 17.8 | 1.217 | 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 | 8.83 E-3 | 126 | 3.05 E-3 | 3.03 E-3 | 34.5 | 1.528 | 1.000 |
libgromacs_mpi.so.7:0x7a70c2 | gmx::set_lincs(InteractionDefinitions const&, int, gmx::Arra... | lincs.cpp:1695 | 0 | 0 | runtime | parallel | 52 | 8.56 E-3 | 127 | 3.60 E-3 | 3.58 E-3 | 42.1 | 1.728 | 1.000 |
libgromacs_mpi.so.7:0x317f15 | Nbnxm::GridSet::putOnGrid(float const (*) [3], int, float co... | gridset.cpp:212 | 0 | 0 | runtime | parallel | 52 | 7.88 E-3 | 127 | 3.56 E-3 | 3.54 E-3 | 45.2 | 1.824 | 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 | 5.20 E-3 | 102 | 2.72 E-3 | 2.64 E-3 | 52.3 | 2.097 | 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 | 3.27 E-3 | 102 | 2.33 E-3 | 2.26 E-3 | 71.3 | 3.488 | 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 | 1.70 E-3 | 2.00 | 862 E-6 | 862 E-6 | 50.6 | 2.024 | 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 | 1.42 E-3 | 2.00 | 648 E-6 | 647 E-6 | 45.5 | 1.836 | 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 | 166 E-6 | 1.00 | 130 E-6 | 130 E-6 | 78.6 | 4.678 | 1.000 |
libgromacs_mpi.so.7:0x4036aa | gmx_ekindata_t::gmx_ekindata_t(int, float, int) | group.cpp:66 | 0 | 0 | runtime | parallel | 52 | 152 E-6 | 1.00 | 132 E-6 | 131 E-6 | 86.4 | 7.332 | 1.000 |
libgromacs_mpi.so.7:0x877f0b | PmeAtomComm& std::vector<PmeAtomComm, std::allocator<PmeAtom... | vector.tcc:480 | 0 | 0 | runtime | parallel | 52 | 139 E-6 | 1.00 | 107 E-6 | 107 E-6 | 77.1 | 4.363 | 1.000 |